HCI Bibliography Home | HCI Conferences | RecSys Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
RecSys Tables of Contents: 070809101112131415

Proceedings of the 2013 ACM Conference on Recommender Systems

Fullname:Proceedings of the 7th ACM conference on Recommender systems
Editors:Qiang Yang; Irwin King; Qing Li; Pearl Pu; George Karypis
Location:Hong Kong, China
Dates:2013-Oct-12 to 2013-Oct-16
Standard No:ISBN: 978-1-4503-2409-0; ACM DL: Table of Contents; hcibib: RecSys13
Links:Conference Website
  1. Technical session: context-aware
  2. Technical session: methods, algorithms, and theory I
  3. Technical session: social media and recommender systems
  4. Technical session: media recommendation
  5. Technical session: user experience
  6. Technical session: beyond ratings
  7. Technical session: methods, algorithms, and theory II
  8. Technical session: scalability
  9. Industry session
  10. Poster session
  11. Doctoral symposium
  12. Demonstrations
  13. Workshops
  14. Tutorials

Technical session: context-aware

Context-aware review helpfulness rating prediction BIBAFull-Text 1-8
  Jiliang Tang; Huiji Gao; Xia Hu; Huan Liu
Online reviews play a vital role in the decision-making process for online users. Helpful reviews are usually buried in a large number of unhelpful reviews, and with the consistently increasing number of reviews, it becomes more and more difficult for online users to find helpful reviews. Therefore most online review websites allow online users to rate the helpfulness of a review and a global helpfulness score is computed for the review based on its available ratings. However, in reality, user-specified helpfulness ratings for reviews are very sparse -- a few reviews attract large numbers of helpfulness ratings while most reviews obtain few or even no helpfulness ratings. The available helpfulness ratings are too sparse for online users to assess the helpfulness of reviews. Also the helpfulness of a review is not necessarily equally useful for all users and users with different background may treat the helpfulness of a review very differently. The user idiosyncracy of review helpfulness motivates us to study the problem of review helpfulness rating prediction in this paper. We first identify various types of context information, model them mathematically, and propose a context-aware review helpfulness rating prediction framework CAP. Experimental results demonstrate the effectiveness of the proposed framework and the importance of context awareness in solving the review helpfulness rating prediction problem.
Query-driven context aware recommendation BIBAFull-Text 9-16
  Negar Hariri; Bamshad Mobasher; Robin Burke
Context aware recommender systems go beyond the traditional personalized recommendation models by incorporating a form of situational awareness. They provide recommendations that not only correspond to a user's preference profile, but that are also tailored to a given situation or context. We consider the setting in which contextual information is represented as a subset of an item feature space describing short-term interests or needs of a user in a given situation. This contextual information can be provided by the user in the form of an explicit query, or derived implicitly.
   We propose a unified probabilistic model that integrates user profiles, item representations, and contextual information. The resulting recommendation framework computes the conditional probability of each item given the user profile and the additional context. These probabilities are used as recommendation scores for ranking items. Our model is an extension of the Latent Dirichlet Allocation (LDA) model that provides the capability for joint modeling of users, items, and the meta-data associated with contexts. Each user profile is modeled as a mixture of the latent topics. The discovered latent topics enable our system to handle missing data in item features. We demonstrate the application of our framework for article and music recommendation. In the latter case, the set of popular tags from social tagging Web sites are used for context descriptions. Our evaluation results show that considering context can help improve the quality of recommendations.
Location-aware music recommendation using auto-tagging and hybrid matching BIBAFull-Text 17-24
  Marius Kaminskas; Francesco Ricci; Markus Schedl
We propose a novel approach to context-aware music recommendation -- recommending music suited for places of interest (POIs). The suggested hybrid approach combines two techniques -- one based on representing both POIs and music with tags, and the other based on the knowledge of the semantic relations between the two types of items. We show that our approach can be scaled up using a novel music auto-tagging technique and we compare it in a live user study to: two non-hybrid solutions, either based on tags or on semantic relations; and to a context-free but personalized recommendation approach. In the considered scenario, i.e., a situation defined by a context (the POI), we show that personalization (via music preference) is not sufficient and it is important to implement effective adaptation techniques to the user's context. In fact, we show that the users are more satisfied with the recommendations generated by combining the tag-based and knowledge-based context adaptation techniques, which exploit orthogonal types of relations between places and music tracks.
Spatial topic modeling in online social media for location recommendation BIBAFull-Text 25-32
  Bo Hu; Martin Ester
Mobile networks enable users to post on social media services (e.g., Twitter) from anywhere. The activities of mobile users involve three major entities: user, post, and location. The interaction of these entities is the key to answer questions such as who will post a message where and on what topic? In this paper, we address the problem of profiling mobile users by modeling their activities, i.e., we explore topic modeling considering the spatial and textual aspects of user posts, and predict future user locations. We propose the first ST (Spatial Topic) model to capture the correlation between users' movements and between user interests and the function of locations. We employ the sparse coding technique which greatly speeds up the learning process. We perform experiments on two real life data sets from Twitter and Yelp. Through comprehensive experiments, we demonstrate that our proposed model consistently improves the average precision@1,5,10,15,20 for location recommendation by at least 50% (Twitter) and 300% (Yelp) against existing state-of-the-art recommendation algorithms and geographical topic models.

Technical session: methods, algorithms, and theory I

Orthogonal query recommendation BIBAFull-Text 33-40
  Hossein Vahabi; Margareta Ackerman; David Loker; Ricardo Baeza-Yates; Alejandro Lopez-Ortiz
One important challenge of current search engines is to satisfy the users' needs when they provide a poorly formulated query. When the pages matching the user's original keywords are judged to be unsatisfactory, query recommendation techniques are used to propose alternative queries and alter the result set. These techniques search for queries that are semantically similar to the user's original query, often searching for keywords that are similar to the keywords given by the user. However, when the original query is sufficiently ill-posed, the user's informational need is best met using entirely different keywords, and a substantially different query may be necessary.
   We propose a novel approach that is not based on the keywords of the original query. We intentionally seek out orthogonal queries, which are related queries that have (almost) no common terms with the user's query. This allows an orthogonal query to satisfy the user's informational need when small perturbations of the original keyword set are insufficient. By using this technique to generate query recommendations, we outperform several known approaches, being the best for long tail queries.
Understanding and improving relational matrix factorization in recommender systems BIBAFull-Text 41-48
  Li Pu; Boi Faltings
Matrix factorization techniques such as the singular value decomposition (SVD) have had great success in recommender systems. We present a new perspective of SVD for constructing a latent space from the training data, which is justified by the theory of hypergraph model. We show that the vectors representing the items in the latent space can be grouped into (approximately) orthogonal clusters which correspond to the vertex clusters in the co-rating hypergraph, and the lengths of the vectors are indicators of the representativeness of the items. These properties are used for making top-$N$ recommendations in a two-phase algorithm. In this work, we provide a new explanation for the significantly better performance of the asymmetric SVD approaches and a novel algorithm for better diversity in top-N recommendations.
Retargeted matrix factorization for collaborative filtering BIBAFull-Text 49-56
  Oluwasanmi Koyejo; Sreangsu Acharyya; Joydeep Ghosh
This paper introduces retargeted matrix factorization (R-MF); a novel approach for learning the user-wise ranking of items in the context of collaborative filtering. R-MF learns to rank by "retargeting" the item ratings of each user, searching for a monotonic transformation of the ratings that results in a better fit while preserving the ranked order of each user's ratings. The retargeting is combined with an underlying matrix factorization regression model that couples the user-wise rankings to exploit shared low dimensional structure. We show that R-MF recovers a unique solution under mild conditions, and propose a simple and efficient optimization scheme that alternates between retargeting the ratings subject to ordering constraints, and matrix factorization regression. The retargeting step is independent for each user, and is trivially parallelized. The ranking performance of retargeted matrix factorization is evaluated on benchmark movie recommendation datasets and results in superior ranking performance compared to collaborative filtering algorithms specifically designed to optimize ranking metrics.
Trading-off among accuracy, similarity, diversity, and long-tail: a graph-based recommendation approach BIBAFull-Text 57-64
  Lei Shi
Improving recommendation accuracy is the mostly focused target of recommendation systems, while it has been increasingly recognized that accuracy is not enough as the only quality criterion. More concepts have been proposed recently to augment the evaluation dimensions, such as similarity, diversity, long-tail, etc. Simultaneously considering multiple criteria leads to a multi-task recommendation. In this paper, a graph-based recommendation approach is proposed to effectively and flexibly trade-off among them.
   Our approach is considered based a 1st order Markovian graph with transition probabilities between user-item pairs. A "cost flow" concept is proposed over the graph, so that items with lower costs are stronger recommended to a user. The cost flows are formulated in a recursive dynamic form, whose stability is proved to be guaranteed by appropriately lower-bounding the transition costs. Furthermore, a mixture of transition costs is designed by combining three ingredients related to long-tail, focusing degree and similarity. To evaluate the ingredients, we propose an orthogonal-sparse-orthogonal nonnegative matrix tri-factorization model and an efficient multiplicative algorithm. Empirical experiments on real-world data show promising results of our approach, which could be regarded as a general framework for other affects if transition costs are designed in various ways.
Nonlinear latent factorization by embedding multiple user interests BIBAFull-Text 65-68
  Jason Weston; Ron J. Weiss; Hector Yee
Classical matrix factorization approaches to collaborative filtering learn a latent vector for each user and each item, and recommendations are scored via the similarity between two such vectors, which are of the same dimension. In this work, we are motivated by the intuition that a user is a much more complicated entity than any single item, and cannot be well described by the same representation. Hence, the variety of a user's interests could be better captured by a more complex representation. We propose to model the user with a richer set of functions, specifically via a set of latent vectors, where each vector captures one of the user's latent interests or tastes. The overall recommendation model is then nonlinear where the matching score between a user and a given item is the maximum matching score over each of the user's latent interests with respect to the item's latent representation. We describe a simple, general and efficient algorithm for learning such a model, and apply it to large scale, real-world datasets from YouTube and Google Music, where our approach outperforms existing techniques.

Technical session: social media and recommender systems

Diffusion-aware personalized social update recommendation BIBAFull-Text 69-76
  Ye Pan; Feng Cong; Kailong Chen; Yong Yu
Many Internet users have encountered serious information overload problem on social networks such as Facebook and Twitter, where users can consume the streams of social updates from their social connections. Traditional methods solving this problem include collaborative filtering and information diffusion modeling. Both methods answer the "who will adopt what" question from different perspective, while either of them only captures single-faceted knowledge of evidences. In this paper, we solve the personalized social update recommendation problem by proposing a framework which integrates the advantages of collaborative filtering and the characteristics of diffusion processes. The main contributions of this paper are three folds. First, we propose a plenty of diffusion features which capture the characteristics of diffusion processes. Second, we build a joint model which takes the advantages of both collaborative filtering and the characteristics of diffusion processes for recommendation. Finally, experiments on two real-world datasets show that our joint model outperforms the methods capturing single-faceted knowledge and several other baselines.
Recommending branded products from social media BIBAFull-Text 77-84
  Yongzheng Zhang; Marco Pennacchiotti
E-commerce companies are increasingly encouraging their users to connect to social media venues such as Facebook and Pinterest. The main strategic goal of such social connections is to boost user interaction and adoption on social media. However only a few efforts have been focused so far on leveraging users' social profiles to personalize the e-commerce experience and to recommend products of interest. In this paper, we start exploring this topic by investigating if a user's social media profile can be used to predict and recommend what type of products and what brands the social user is more likely to buy. More specifically, we study the correlation between the brands liked by the user on social media sites and those purchased on an e-commerce site. We then leverage these correlations in a brand prediction system, showing that social media can be effectively used to recommend branded products when user-user collaborative filtering techniques are used.
Top-N recommendations from implicit feedback leveraging linked open data BIBAFull-Text 85-92
  Vito Claudio Ostuni; Tommaso Di Noia; Eugenio Di Sciascio; Roberto Mirizzi
The advent of the Linked Open Data (LOD) initiative gave birth to a variety of open knowledge bases freely accessible on the Web. They provide a valuable source of information that can improve conventional recommender systems, if properly exploited. In this paper we present SPrank, a novel hybrid recommendation algorithm able to compute top-N item recommendations from implicit feedback exploiting the information available in the so called Web of Data. We leverage DBpedia, a well-known knowledge base in the LOD compass, to extract semantic path-based features and to eventually compute recommendations using a learning to rank algorithm. Experiments with datasets on two different domains show that the proposed approach outperforms in terms of prediction accuracy several state-of-the-art top-N recommendation algorithms for implicit feedback in situations affected by different degrees of data sparsity.
Exploring temporal effects for location recommendation on location-based social networks BIBAFull-Text 93-100
  Huiji Gao; Jiliang Tang; Xia Hu; Huan Liu
Location-based social networks (LBSNs) have attracted an inordinate number of users and greatly enriched the urban experience in recent years. The availability of spatial, temporal and social information in online LBSNs offers an unprecedented opportunity to study various aspects of human behavior, and enable a variety of location-based services such as location recommendation. Previous work studied spatial and social influences on location recommendation in LBSNs. Due to the strong correlations between a user's check-in time and the corresponding check-in location, recommender systems designed for location recommendation inevitably need to consider temporal effects. In this paper, we introduce a novel location recommendation framework, based on the temporal properties of user movement observed from a real-world LBSN dataset. The experimental results exhibit the significance of temporal patterns in explaining user behavior, and demonstrate their power to improve location recommendation performance.
The curated web: a recommendation challenge BIBAFull-Text 101-104
  Zurina Saaya; Rachael Rafter; Markus Schaal; Barry Smyth
In this paper we consider the application of content-based recommendation techniques to web curation services which allow users to curate and share topical collections of content (e.g. images, news, web pages etc.). Curation services like Pinterest are now a mainstay of the modern web and present a range of interesting recommendation challenges. In this paper we consider the task of recommending collections to users and evaluate a range of different content-based techniques across a variety of content signals. We present the results of a large-scale evaluation using data from the Scoop.it web page curation service.

Technical session: media recommendation

Personalized news recommendation with context trees BIBAFull-Text 105-112
  Florent Garcin; Christos Dimitrakakis; Boi Faltings
The proliferation of online news creates a need for filtering interesting articles. Compared to other products, however, recommending news has specific challenges: news preferences are subject to trends, users do not want to see multiple articles with similar content, and frequently we have insufficient information to profile the reader.
   In this paper, we introduce a class of news recommendation systems based on context trees. They can provide high-quality news recommendations to anonymous visitors based on present browsing behaviour. Using an unbiased testing methodology, we show that they make accurate and novel recommendations, and that they are sufficiently flexible for the challenges of news recommendation.
What to read next?: making personalized book recommendations for K-12 users BIBAFull-Text 113-120
  Maria Soledad Pera; Yiu-Kai Ng
Finding books that children/teenagers are interested in these days is a non-trivial task due to the diversity of topics covered in huge volumes of books with varied readability levels. Even though K-12 readers can turn to book recommenders to look for books, the recommended books may not satisfy their personal needs, since they could be beyond/below their readability levels or fail to match their topics of interest. To address these problems, we introduce BReK12, a book recommender that makes personalized suggestions tailored to each K-12 user U based on books available on a social book-marking site that (i) are similar in content to the ones that are known to be of interest to U, (ii) have been bookmarked by users with reading patterns similar to U's, and (iii) can be comprehended by U. BReK12 is an asset to its users, since it suggests books that are appealing to its users and at grade levels that they can cope with, which can increase their reading selection choices and motivate them to read. We have also developed ReLAT, the readability analysis tool employed by BReK12 to determine the grade level of books. ReLAT is novel, compared with existing readability formulas, since it can predict the grade level of a book even if an excerpt of the book is not available. We have conducted empirical studies which have verified the accuracy of ReLAT in predicting the grade level of a book and the effectiveness of BReK12 over existing baseline recommendation systems.
Movie recommender system for profit maximization BIBAFull-Text 121-128
  Amos Azaria; Avinatan Hassidim; Sarit Kraus; Adi Eshkol; Ofer Weintraub; Irit Netanely
Traditional recommender systems minimize prediction error with respect to users' choices. Recent studies have shown that recommender systems have a positive effect on the provider's revenue.
   In this paper we show that by providing a set of recommendations different than the one perceived best according to user acceptance rate, the recommendation system can further increase the business' utility (e.g. revenue), without any significant drop in user satisfaction. Indeed, the recommendation system designer should have in mind both the user, whose taste we need to reveal, and the business, which wants to promote specific content.
   We performed a large body of experiments comparing a commercial state-of-the-art recommendation engine with a modified recommendation list, which takes into account the utility (or revenue) which the business obtains from each suggestion that is accepted by the user. We show that the modified recommendation list is more desirable for the business, as the end result gives the business a higher utility (or revenue). To study possible reduce in satisfaction by providing the user worse suggestions, we asked the users how they perceive the list of recommendation that they received. Differences in user satisfaction between the lists is negligible, and not statistically significant.
   We also uncover a phenomenon where movie consumers prefer watching and even paying for movies that they have already seen in the past than movies that are new to them.
Xbox movies recommendations: variational Bayes matrix factorization with embedded feature selection BIBAFull-Text 129-136
  Noam Koenigstein; Ulrich Paquet
We present a matrix factorization model inspired by challenges we encountered while working on the Xbox movies recommendation system. The item catalog in a recommender system is typically equipped with meta-data features in the form of labels. However, only part of these features are informative or useful with regard to collaborative filtering. By incorporating a novel sparsity prior on feature parameters, the model automatically discerns and utilizes informative features while simultaneously pruning non-informative features.
   The model is designed for binary feedback, which is common in many real-world systems where numeric rating data is scarce or non-existent. However, the overall framework is applicable to any likelihood function. Model parameters are estimated with a Variational Bayes inference algorithm, which is robust to over-fitting and does not require cross-validation and fine tuning of regularization coefficients. The efficacy of our method is illustrated on a sample from the Xbox movies dataset as well as on the publicly available MovieLens dataset. In both cases, the proposed solution provides superior predictive accuracy, especially for long-tail items. We then demonstrate the feature selection capabilities and compare against the common case of simple Gaussian priors. Finally, we show that even without features, our model performs better than a baseline model trained with the popular stochastic gradient descent approach.
Personalized next-song recommendation in online karaokes BIBAFull-Text 137-140
  Xiang Wu; Qi Liu; Enhong Chen; Liang He; Jingsong Lv; Can Cao; Guoping Hu
In this paper, we propose Personalized Markov Embedding (PME), a next-song recommendation strategy for online karaoke users. By modeling the sequential singing behavior, we first embed songs and users into a Euclidean space in which distances between songs and users reflect the strength of their relationships. Then, given each user's last song, we can generate personalized recommendations by ranking the candidate songs according to the embedding. Moreover, PME can be trained without any requirement of content information. Finally, we perform an experimental evaluation on a real world data set provided by ihou.com which is an online karaoke website launched by iFLYTEK, and the results clearly demonstrate the effectiveness of PME.

Technical session: user experience

Topic diversity in tag recommendation BIBAFull-Text 141-148
  Fabiano Belém; Rodrygo Santos; Jussara Almeida; Marcos Gonçalves
Tag recommendation approaches have historically focused on maximizing the relevance of the recommended tags for a given object, such as a movie or a song. Nevertheless, different users may be interested in the same object for different reasons -- for instance, the Star Wars movies may appeal to both adventure as well as to fantasy movie fans. In this situation, a sensible strategy is to provide a user with diverse recommendations of how to tag the object. In this paper, we address the problem of recommending relevant and diverse tags as a ranking problem. In particular, we propose a novel tag recommendation approach that explicitly takes into account the possible topics (e.g., categories) underlying an object in order to promote tags with high coverage and low redundancy with respect to these topics. We thoroughly evaluate our proposed approach using data collected from two popular Web 2.0 applications, namely, LastFM and MovieLens. Our experimental results attest the effectiveness of our approach at promoting more relevant and diverse tags in contrast to state-of-the-art relevance-based methods as well as a recently proposed method that takes both relevance and diversity into account.
Rating support interfaces to improve user experience and recommender accuracy BIBAFull-Text 149-156
  Tien T. Nguyen; Daniel Kluver; Ting-Yu Wang; Pik-Mai Hui; Michael D. Ekstrand; Martijn C. Willemsen; John Riedl
One of the challenges for recommender systems is that users struggle to accurately map their internal preferences to external measures of quality such as ratings. We study two methods for supporting the mapping process: (i) reminding the user of characteristics of items by providing personalized tags and (ii) relating rating decisions to prior rating decisions using exemplars. In our study, we introduce interfaces that provide these methods of support. We also present a set of methodologies to evaluate the efficacy of the new interfaces via a user experiment. Our results suggest that presenting exemplars during the rating process helps users rate more consistently, and increases the quality of the data.
ReComment: towards critiquing-based recommendation with speech interaction BIBAFull-Text 157-164
  Peter Grasch; Alexander Felfernig; Florian Reinfrank
In contrast to search-based approaches, critiquing-based recommender systems provide a navigation-based interface where users are enabled to critique displayed recommendations as a means of preference elicitation. In this paper we present ReComment, our approach to natural language based unit critiquing. We discuss the developed prototype and present the corresponding user interface. In order to show the applicability of our concepts, we present the results of a user study. This study shows that speech interfaces have the potential to improve the perceived ease of use as well as the overall quality of recommendations.
Hidden factors and hidden topics: understanding rating dimensions with review text BIBAFull-Text 165-172
  Julian McAuley; Jure Leskovec
In order to recommend products to users we must ultimately predict how a user will respond to a new product. To do so we must uncover the implicit tastes of each user as well as the properties of each product. For example, in order to predict whether a user will enjoy Harry Potter, it helps to identify that the book is about wizards, as well as the user's level of interest in wizardry. User feedback is required to discover these latent product and user dimensions. Such feedback often comes in the form of a numeric rating accompanied by review text. However, traditional methods often discard review text, which makes user and product latent dimensions difficult to interpret, since they ignore the very text that justifies a user's rating. In this paper, we aim to combine latent rating dimensions (such as those of latent-factor recommender systems) with latent review topics (such as those learned by topic models like LDA). Our approach has several advantages. Firstly, we obtain highly interpretable textual labels for latent rating dimensions, which helps us to 'justify' ratings with text. Secondly, our approach more accurately predicts product ratings by harnessing the information present in review text; this is especially true for new products and users, who may have too few ratings to model their latent factors, yet may still provide substantial information from the text of even a single review. Thirdly, our discovered topics can be used to facilitate other tasks such as automated genre discovery, and to identify useful and representative reviews.
Improving augmented reality using recommender systems BIBAFull-Text 173-176
  Zhuo Zhang; Shang Shang; Sanjeev R. Kulkarni; Pan Hui
With the rapid development of smart devices and wireless communication, especially with the pre-launch of Google Glass, augmented reality (AR) has received enormous attention recently. AR adds virtual objects into a user's real-world environment enabling live interaction in three dimensions. Limited by the small display of AR devices, content selection is one of the key issues to improve user experience. In this paper, we present an aggregated random walk algorithm incorporating personal preferences, location information, and temporal information in a layered graph. By adaptively changing the graph edge weight and computing the rank score, the proposed AR recommender system predicts users' preferences and provides the most relevant recommendations with aggregated information.

Technical session: beyond ratings

Exploiting non-content preference attributes through hybrid recommendation method BIBAFull-Text 177-184
  Fernando Mourão; Leonardo Rocha; Joseph A. Konstan; Wagner, Jr. Meira
This paper explores a method for incorporating into a recommender system explicit representations of user's preferences over non-content attributes such as popularity, recency, and similarity of recommended items. We show how such attributes can be modeled as a preference vector that can be used in a vector-space content-based recommender, and how that content-based recommender can be integrated with various collaborative filtering techniques through re-weighting of Top-M recommendations. We evaluate this approach on several recommender systems datasets and collaborative filtering methods, and find that incorporating the three preference attributes can lead to a substantial increase in Top-50 precision while also enhancing diversity and novelty.
Hybrid event recommendation using linked data and user diversity BIBAFull-Text 185-192
  Houda Khrouf; Raphaël Troncy
An ever increasing number of social services offer thousands of diverse events per day. Users tend to be overwhelmed by the massive amount of information available, especially with limited browsing options perceived in many event web services. To alleviate this information overload, a recommender system becomes a vital component for assisting users selecting relevant events. However, such system faces a number of challenges owed to the inherent complex nature of an event. In this paper, we propose a novel hybrid approach built on top of Semantic Web. On the one hand, we use a content-based system enriched with Linked Data to overcome the data sparsity, a problem induced by the transiency of events. On the other hand, we incorporate a collaborative filtering to involve the social aspect, an influential feature in decision making. This hybrid system is enhanced by the integration of a user diversity model designed to detect user propensity towards specific topics. We show how the hybridization of CB+CF systems and the integration of interest diversity features are important to improve predictions. Experimental results demonstrate the effectiveness of our approach using precision and recall measures.
Pairwise learning in recommendation: experiments with community recommendation on LinkedIn BIBAFull-Text 193-200
  Amit Sharma; Baoshi Yan
Many online systems present a list of recommendations and infer user interests implicitly from clicks or other contextual actions. For modeling user feedback in such settings, a common approach is to consider items acted upon to be relevant to the user, and irrelevant otherwise. However, clicking some but not others conveys an implicit ordering of the presented items. Pairwise learning, which leverages such implicit ordering between a pair of items, has been successful in areas such as search ranking. In this work, we study whether pairwise learning can improve community recommendation. We first present two novel pairwise models adapted from logistic regression. Both offline and online experiments in a large real-world setting show that incorporating pairwise learning improves the recommendation performance. However, the improvement is only slight. We find that users' preferences regarding the kinds of communities they like can differ greatly, which adversely affect the effectiveness of features derived from pairwise comparisons. We therefore propose a probabilistic latent semantic indexing model for pairwise learning (Pairwise PLSI), which assumes a set of users' latent preferences between pairs of items. Our experiments show favorable results for the Pairwise PLSI model and point to the potential of using pairwise learning for community recommendation.
Which app will you use next?: collaborative filtering with interactional context BIBAFull-Text 201-208
  Nagarajan Natarajan; Donghyuk Shin; Inderjit S. Dhillon
The application a smart phone user will launch next intuitively depends on the sequence of apps used recently. More generally, when users interact with systems such as shopping websites or online radio, they click on items that are of interest in the current context. We call the sequence of clicks made in the current session interactional context. It is desirable for a recommender system to use the context set by the user to update recommendations. Most current context-aware recommender systems focus on a relatively less dynamic representational context defined by attributes such as season, location and tastes. In this paper, we study the problem of collaborative filtering with interactional context, where the goal is to make personalized and dynamic recommendations to a user engaged in a session. To this end, we propose the methodname algorithm that works in two stages. First, users are clustered by their transition behavior (one-step Markov transition probabilities between items), and cluster-level Markov models are computed. Then personalized PageRank is computed for a given user on the corresponding cluster Markov graph, with a personalization vector derived from the current context. We give an interpretation of the second stage of the algorithm as adding an appropriate context bias, in addition to click bias (or rating bias), to a classical neighborhood-based collaborative filtering model, where the neighborhood is determined from a Markov graph. Experimental results on two real-life datasets demonstrate the superior performance of our algorithm, where we achieve at least 20% (up to 37%) improvement over competitive methods in the recall level at top-20.
A food recommender for patients in a care facility BIBAFull-Text 209-212
  Toon De Pessemier; Simon Dooms; Luc Martens
In this research, a food recommendation strategy for patients in a care facility is proposed. Since many of these patients cannot express their personal preferences, a recommender system can assist the caregivers in the selection of the menu items that match the patients' preferences. Recommendations are generated based on three information sources: explicit ratings for menu items, implicit feedback based on the patient's eating behavior and the amount of food that was eaten, and inferred preferences for the ingredients of the menu items. In addition, monitoring the amount of food that was eaten by each patient can provide insights into the optimal amount of each menu item that has to be served to each patient. Furthermore, monitoring food consumption allows to detect irregularities in the eating behavior of the patient, which can be a symptom of illness.

Technical session: methods, algorithms, and theory II

Evaluation of recommendations: rating-prediction and ranking BIBAFull-Text 213-220
  Harald Steck
The literature on recommender systems distinguishes typically between two broad categories of measuring recommendation accuracy: rating prediction, often quantified in terms of the root mean square error (RMSE), and ranking, measured in terms of metrics like precision and recall, among others. In this paper, we examine both approaches in detail, and find that the dominating difference lies instead in the training and test data considered: rating prediction is concerned with only the observed ratings, while ranking typically accounts for all items in the collection, whether the user has rated them or not. Furthermore, we show that predicting observed ratings, while popular in the literature, only solves a (small) part of the rating prediction task for any item in the collection, which is a common real-world problem. The reasons are selection bias in the data, combined with data sparsity. We show that the latter rating-prediction task involves the prediction task 'Who rated What' as a sub-problem, which can be cast as a classification or ranking problem. This suggests that solving the ranking problem is not only valuable by itself, but also for predicting the rating value of any item.
You are what you consume: a Bayesian method for personalized recommendations BIBAFull-Text 221-228
  Konstantinos Babas; Georgios Chalkiadakis; Evangelos Tripolitakis
In this paper, we propose a novel Bayesian approach for personalized recommendations. In our approach, we model both user preferences and items under recommendation as multivariate Gaussian distributions; and make use of Normal-Inverse Wishart priors to model the recommendation agent beliefs about user types. We employ a lightweight agent-user interaction process, during which the user is presented with and asked to rate a small number of items. We then interpret these ratings in an innovative way, using them to guide a Bayesian updating process that helps us both capture a user's current mood, and maintain her overall user type. We produced several variants of our approach, and applied them in the movie recommendations domain, evaluating them on data from the MovieLens dataset. Our algorithms are shown to be competitive against a state-of-the-art method, which nevertheless requires a minimum set of ratings from various users to provide recommendations -- unlike our entirely personalized approach.
To personalize or not: a risk management perspective BIBAFull-Text 229-236
  Weinan Zhang; Jun Wang; Bowei Chen; Xiaoxue Zhao
Personalization techniques have been widely adopted in many recommender systems. However, experiments on real-world datasets show that for some users in certain contexts, personalized recommendations do not necessarily perform better than recommendations that rely purely on popularity. Broadly, this can be interpreted by the fact that the parameters of a personalization model are usually estimated from sparse data; the resulting personalized prediction, despite of its low bias, is often volatile. In this paper, we study the problem further by investigating into the ranking of recommendation lists. From a risk management and portfolio retrieval perspective, there is no difference between the popularity-based and the personalized ranking as both of the recommendation outputs can be represented as the trade-off between expected relevance (reward) and associated uncertainty (risk). Through our analysis, we discover common scenarios and provide a technique to predict whether personalization will fail. Besides the theoretical understanding, our experimental results show that the resulting switch algorithm, which decides whether or not to personalize, outperforms the mainstream recommendation algorithms.
Online multi-task collaborative filtering for on-the-fly recommender systems BIBAFull-Text 237-244
  Jialei Wang; Steven C. H. Hoi; Peilin Zhao; Zhi-Yong Liu
Traditional batch model-based Collaborative Filtering (CF) approaches typically assume a collection of users' rating data is given a priori for training the model. They suffer from a common yet critical drawback, i.e., the model has to be re-trained completely from scratch whenever new training data arrives, which is clearly non-scalable for large real recommender systems where users' rating data often arrives sequentially and frequently. In this paper, we investigate a novel efficient and scalable online collaborative filtering technique for on-the-fly recommender systems, which is able to effectively online update the recommendation model from a sequence of rating observations. Specifically, we propose a family of online multi-task collaborative filtering (OMTCF) algorithms, which tackle the online collaborative filtering task by exploiting the similar principle as online multitask learning. Encouraging empirical results on large-scale datasets showed that the proposed technique is significantly more effective than the state-of-the-art algorithms.
Learning to rank recommendations with the k-order statistic loss BIBAFull-Text 245-248
  Jason Weston; Hector Yee; Ron J. Weiss
Making recommendations by learning to rank is becoming an increasingly studied area. Approaches that use stochastic gradient descent scale well to large collaborative filtering datasets, and it has been shown how to approximately optimize the mean rank, or more recently the top of the ranked list. In this work we present a family of loss functions, the k-order statistic loss, that includes these previous approaches as special cases, and also derives new ones that we show to be useful. In particular, we present (i) a new variant that more accurately optimizes precision at k, and (ii) a novel procedure of optimizing the mean maximum rank, which we hypothesize is useful to more accurately cover all of the user's tastes. The general approach works by sampling N positive items, ordering them by the score assigned by the model, and then weighting the example as a function of this ordered set. Our approach is studied in two real-world systems, Google Music and YouTube video recommendations, where we obtain improvements for computable metrics, and in the YouTube case, increased user click through and watch duration when deployed live on www.youtube.com.

Technical session: scalability

A fast parallel SGD for matrix factorization in shared memory systems BIBAFull-Text 249-256
  Yong Zhuang; Wei-Sheng Chin; Yu-Chin Juan; Chih-Jen Lin
Matrix factorization is known to be an effective method for recommender systems that are given only the ratings from users to items. Currently, stochastic gradient descent (SGD) is one of the most popular algorithms for matrix factorization. However, as a sequential approach, SGD is difficult to be parallelized for handling web-scale problems. In this paper, we develop a fast parallel SGD method, FPSGD, for shared memory systems. By dramatically reducing the cache-miss rate and carefully addressing the load balance of threads, FPSGD is more efficient than state-of-the-art parallel algorithms for matrix factorization.
DrunkardMob: billions of random walks on just a PC BIBAFull-Text 257-264
  Aapo Kyrola
Random walks on graphs are a staple of many ranking and recommendation algorithms. Simulating random walks on a graph which fits in memory is trivial, but massive graphs pose a problem: the latency of following walks across network in a cluster or loading nodes from disk on-demand renders basic random walk simulation unbearably inefficient. In this work we propose DrunkardMob, a new algorithm for simulating hundreds of millions, or even billions, of random walks on massive graphs, on just a single PC or laptop. Instead of simulating one walk a time it processes millions of them in parallel, in a batch. Based on DrunkardMob and GraphChi we further propose a framework for easily expressing scalable algorithms based on graph walks.
Using maximum coverage to optimize recommendation systems in e-commerce BIBAFull-Text 265-272
  Mikael Hammar; Robin Karlsson; Bengt J. Nilsson
We study the problem of optimizing recommendation systems for e-commerce sites. We consider in particular a combinatorial solution to this optimization based on the well known Maximum Coverage problem that asks for the k sets (products) that cover the most elements from a ground set (consumers). This formulation provides an abstract model for what k products should be recommended to maximize the probability of consumer purchase. Unfortunately, Maximum Coverage is NP-complete but an efficient approximation algorithm exists based on the Greedy methodology.
   We exhibit test results from the Greedy method on real data sets showing 3-8% increase in sales using the Maximum Coverage optimization method in comparison to the standard best-seller list. A secondary effect that our Greedy algorithm exhibits on the tested data is increased diversification in presented products over the best-seller list.
Efficient top-n recommendation for very large scale binary rated datasets BIBAFull-Text 273-280
  Fabio Aiolli
We present a simple and scalable algorithm for top-N recommendation able to deal with very large datasets and (binary rated) implicit feedback. We focus on memory-based collaborative filtering algorithms similar to the well known neighbour based technique for explicit feedback. The major difference, that makes the algorithm particularly scalable, is that it uses positive feedback only and no explicit computation of the complete (user-by-user or item-by-item) similarity matrix needs to be performed.
   The study of the proposed algorithm has been conducted on data from the Million Songs Dataset (MSD) challenge whose task was to suggest a set of songs (out of more than 380k available songs) to more than 100k users given half of the user listening history and complete listening history of other 1 million people.
   In particular, we investigate on the entire recommendation pipeline, starting from the definition of suitable similarity and scoring functions and suggestions on how to aggregate multiple ranking strategies to define the overall recommendation. The technique we are proposing extends and improves the one that already won the MSD challenge last year.
Distributed matrix factorization with mapreduce using a series of broadcast-joins BIBAFull-Text 281-284
  Sebastian Schelter; Christoph Boden; Martin Schenck; Alexander Alexandrov; Volker Markl
The efficient, distributed factorization of large matrices on clusters of commodity machines is crucial to applying latent factor models in industrial-scale recommender systems. We propose an efficient, data-parallel low-rank matrix factorization with Alternating Least Squares which uses a series of broadcast-joins that can be efficiently executed with MapReduce.
   We empirically show that the performance of our solution is suitable for real-world use cases. We present experiments on two publicly available datasets and on a synthetic dataset termed Bigflix, generated from the Netflix dataset. Bigflix contains 25 million users and more than 5 billion ratings, mimicking data sizes recently reported as Netflix' production workload. We demonstrate that our approach is able to run an iteration of Alternating Least Squares in six minutes on this dataset. Our implementation has been contributed to the open source machine learning library Apache Mahout.

Industry session

Catch-up TV recommendations: show old favourites and find new ones BIBAFull-Text 285-294
  Mengxi Xu; Shlomo Berkovsky; Sebastien Ardon; Sipat Triukose; Anirban Mahanti; Irena Koprinska
Web-based catch-up TV has revolutionised watching habits as it provides users the opportunity to watch programs at their preferred time and place, using a variety of devices. With the increasing offer of TV content, there is an emergent need for personalised recommendation solutions, which help users to select programs of interest. In this work, we study the watching patterns of users of an Australian nation-wide catch-up TV service provider and develop a suite of approaches for a catch-up recommendation scenario. We evaluate these approaches using a new large-scale dataset gathered by the Web-based catch-up portal deployed by the provider. The evaluation allows us to compare the performance of several recommenders that address the discovery of both TV programs already watched by users and new programs that users may find relevant.
Generating supplemental content information using virtual profiles BIBAFull-Text 295-302
  Haishan Liu; Mohammad Amin; Baoshi Yan; Anmol Bhasin
We describe a hybrid recommendation system at LinkedIn that seeks to optimally extract relevant information pertaining to items to be recommended. By extending the notion of an item profile, we propose the concept of a "virtual profile" that augments the content of the item with rich set of features inherited from members who have already shown explicit interest in it. Unlike item-based collaborative filtering, we focus on discovering the characteristic descriptors that underlie the item-user association. Such information is used as supplemental features in a content-based filtering system. The main objective of virtual profiles is to provide a means to tap into rich-content information from one type of entity and propagate features extracted from which to other affiliated entities that may suffer from relative data scarcity. We empirically evaluate the proposed method on a real-world community recommendation problem at LinkedIn. The result shows that the virtual profiles outperform a collaborative filtering based approach (user who likes this also likes that). In particular, the improvement is more significant for new users with only limited connections, demonstrating the capability of the method to address the cold-start problem in pure collaborative filtering systems.

Poster session

A people-to-people content-based reciprocal recommender using hidden Markov models BIBAFull-Text 303-306
  Ammar Alanazi; Michael Bain
Users of online social networks such as dating websites often need help to find successful matches. People-to-people recommender systems can be used in social networks to help users find better matches, which requires solving the problem of reciprocal recommendation. However, most existing reciprocal recommenders use either profile similarity or interaction similarity to recommend new matches, without considering temporal features. In this paper we introduce a method for temporal reciprocal recommender systems using Hidden Markov Models to generate recommendations. Instead of summarising the whole historical data in one past state, we propose a model that formalises historical data on interactions as a series of successive states changing over time and then tries to find the recommended next state. We have implemented this new approach and the results of testing on industrial-scale data from a real dating website show a noticeable improvement over the previous best-performing recommenders.
Acquiring user profiles from implicit feedback in a conversational recommender system BIBAFull-Text 307-310
  Henry Blanco; Francesco Ricci
Query revisions in a conversational system can be efficiently computed by assuming that the profiles of the potential users are in a predefined, a priori known and finite set. However, without any additional knowledge of the actual profiles distribution, the system may miss the true profiles of the users, hence deteriorating the system performance. We propose a method for identifying a tailored set of profiles that is acquired by analysing the implicitly shown preferences of the users that interacted with the system. We show that with the proposed method the system can efficiently identify good query revisions.
A system for advice provision in multiple prospect-selection problems BIBAFull-Text 311-314
  Amos Azaria; Sarit Kraus; Ariella Richardson
When humans face a broad spectrum of topics, where each topic consists of several options, they usually make a decision on each topic separately. Usually, a person will perform better by making a global decision, however, taking all consequences into account is extremely difficult. We present a novel computational method for advice-generation in an environment where people need to decide among multiple selection problems. This method is based on the prospect theory and uses machine learning techniques. We graphically present this advice to the users and compare it with advice which encourages the users to always select the option with a higher expected outcome. We show that our method outperforms the expected outcome approach in terms of user and satisfaction.
Clustering-based factorized collaborative filtering BIBAFull-Text 315-318
  Nima Mirbakhsh; Charles X. Ling
Factorized collaborative models show a promising accuracy and scalability in recommendation systems. They employ the latent collaborative information of users and items to achieve higher accuracy of recommendation. In this paper, we propose a new approach to improve the accuracy of two well-known, highly scalable factorized models: SVD++ and Asymmetric-SVD++. These are cutting-edge factorized models that have played a key role in the Netflix prize winner's solution. We first employ collaborative information to categorize the users and items. We then discover the shared interests between these categories. Including this new information, we extend these cutting-edge models regarding two main goals: 1) to improve their recommendation accuracies; 2) to keep the extended models still scalable. Finally, we evaluate our proposed models on two recommendation datasets: MovieLens100k, and Netflix. Our experiment shows that adding the shared interests among categories into these models improves their accuracy while maintaining scalability.
Cross social networks interests predictions based ongraph features BIBAFull-Text 319-322
  Amit Tiroshi; Shlomo Berkovsky; Mohamed Ali Kaafar; Terence Chen; Tsvi Kuflik
The tremendous popularity of Online Social Networks (OSN) has led to situations, where users have their profiles spread across multiple networks. These partial profiles reflect different user characteristics, depending mainly on the nature of the network, e.g., Facebook's social vs. LinkedIn's professional focus. Combining data gathered by multiple networks may benefit individual users, and the community as a whole, as this could facilitate the provision of more accurate services and recommendations. This paper reports on an exploratory study of the process of making such recommendations using a unique multi-network dataset containing user interests across multiple domains, e.g., music, books, and movies. We represent the data using a graph model and generate recommendations using a set of features extracted from and populated by the model. We assess the contribution of various network- and domain-related features to the accuracy of the recommendations and motivate future work into automated feature selection.
Differential data analysis for recommender systems BIBAFull-Text 323-326
  Richard Chow; Hongxia Jin; Bart Knijnenburg; Gokay Saldamli
We present techniques to characterize which data contributes most to the accuracy of a recommendation algorithm. Our main technique is called differential data analysis. The name is inspired by other sorts of differential analysis, such as differential power analysis and differential cryptanalysis, where insight comes through analysis of slightly differing inputs. In differential data analysis we chunk the data and compare results in the presence or absence of each chunk. We apply differential data analysis to two datasets and three different attributes. The first attribute is called user hardship. This is a novel attribute, particularly relevant to location datasets, that indicates how burdensome a data point was to achieve. The second and third attributes are more standard: timestamp and user rating. For user rating, we confirm previous work concerning the increased importance to the recommender of high and low user ratings.
Effectiveness of the data generated on different time in latent factor model BIBAFull-Text 327-330
  Qianru Zheng; Horace H. S. Ip
User selection data accumulates as time goes by. Although the recent selections are usually assumed to have higher impact on the recommendation accuracy, empirical studies on this problem are limited. For old data, whether they can contribute to the recommendation accuracy is still to be determined. On one hand, changes in short-term user preference over time may limit their effectiveness in prediction, but on the other hand, one cannot rule out their potential in capturing long term user preferences. The result is important for the system owner to determine which data is useful to make the recommendation accurately. While there have been some related studies on the time dependency of data quality using neighbor-based CF methods (e.g., [4]), its effects remain unverified for other CF methods. In this paper, we study the effect of data generated over different time period on recommendation precision using several popular model-based CF algorithms (latent factor models). experiment results show that while more recent data expectedly have larger impacts, the usefulness of older data cannot be ignored as long as there are sufficient old samples. However, the addition of insufficient amount of old data seems to have negative impacts.
Interview process learning for top-n recommendation BIBAFull-Text 331-334
  Fangwei Hu; Yong Yu
In the field of recommendation system research, a key challenge is how to effectively recommend items for new users, a problem generally known as cold-start recommendation. In order to alleviate cold-start problem, recently systems try to get the users' interests by progressively querying users' preference on predefined items. Constructing the query process via machine learning based techniques becomes an important direction to solve cold-start problem. In this paper, we propose a novel interview process learning algorithm. Different from previous approaches which focus on rate prediction, our model is able to handle wide ranges of loss functions and can be used in collaborative ranking task. Experimental results on three real world recommendation dataset demonstrate that our proposed method outperforms several baseline methods.
Escape the bubble: guided exploration of music preferences for serendipity and novelty BIBAFull-Text 335-338
  Maria Taramigkou; Efthimios Bothos; Konstantinos Christidis; Dimitris Apostolou; Gregoris Mentzas
In order to predict user behaviour recommender systems generate views of the world according to expressed and known user preferences resulting in 'filter bubbles'. This kind of bubbles generally help users to easily identify objects they like. However, it is becoming increasingly difficult for users to escape their personalized world and change their perspectives especially in domains such as music. In this work we present a methodology and related system that allows users to initiate explorations of music genres by taking a gradual path towards the desired genre while viewing the preferences of other users. The proposed methodology is based on identifying 'latent genres' and using user preference graphs for detecting optimal paths towards a selected target latent genre. In this process we generate suggestions of artists a user should listen to, aiming towards serendipitous and novel encounters. We have implemented our approach in a music recommendation system and evaluated it with encouraging results.
Evaluating top-n recommendations "when the best are gone" BIBAFull-Text 339-342
  Paolo Cremonesi; Franca Garzotto; Massimo Quadrana
In a number of domains of interest for recommender systems, items are characterized by constrained and variable "capacity": the same product or service can be consumed by a limited number of users and the possibility of item consumption depends on contextual circumstances (e.g., time). Our work explores recommenders in the context of these "bounded" domains. We consider online hotel booking as a case study, and investigates if and how "missing" items (hotels that eventually becomes unavailable for users' consumption) affect the quality of recommendations. The paper proposes a technique for defining "missing" items as "best items", and presents an articulated empirical research in which recommendations for hotel online booking are evaluated in different experimental conditions with a user centric approach involving 142 participants.
An analysis of tag-recommender evaluation procedures BIBAFull-Text 343-346
  Stephan Doerfel; Robert Jäschke
Since the rise of collaborative tagging systems on the web, the tag recommendation task -- suggesting suitable tags to users of such systems while they add resources to their collection -- has been tackled. However, the (offline) evaluation of tag recommendation algorithms usually suffers from difficulties like the sparseness of the data or the cold start problem for new resources or users. Previous studies therefore often used so-called post-cores (specific subsets of the original datasets) for their experiments. In this paper, we conduct a large-scale experiment in which we analyze different tag recommendation algorithms on different cores of three real-world datasets. We show, that a recommender's performance depends on the particular core and explore correlations between performances on different cores.
Recommendation in heterogeneous information networks with implicit user feedback BIBAFull-Text 347-350
  Xiao Yu; Xiang Ren; Yizhou Sun; Bradley Sturt; Urvashi Khandelwal; Quanquan Gu; Brandon Norick; Jiawei Han
Recent studies suggest that by using additional user or item relationship information when building hybrid recommender systems, the recommendation quality can be largely improved. However, most such studies only consider a single type of relationship, e.g., social network. Notice that in many applications, the recommendation problem exists in an attribute-rich heterogeneous information network environment. In this paper, we study the entity recommendation problem in heterogeneous information networks. We propose to combine various relationship information from the network with user feedback to provide high quality recommendation results.
   The major challenge of building recommender systems in heterogeneous information networks is to systematically define features to represent the different types of relationships between entities, and learn the importance of each relationship type. In the proposed framework, we first use meta-path-based latent features to represent the connectivity between users and items along different paths in the related information network. We then define a recommendation model with such latent features and use Bayesian ranking optimization techniques to estimate the model. Empirical studies show that our approach outperforms several widely employed implicit feedback entity recommendation techniques.
Recommendation opportunities: improving item prediction using weighted percentile methods in collaborative filtering systems BIBAFull-Text 351-354
  Panagiotis Adamopoulos; Alexander Tuzhilin
This paper proposes a novel method for estimating unknown ratings and recommendation opportunities and illustrates the practical implementation of the proposed approach by presenting a certain variation of the classical k-NN method in neighborhood-based collaborative filtering systems using weighted percentiles. We conduct an empirical study showing that the proposed method outperforms the standard user-based collaborative filtering approach by a wide margin in terms of item prediction accuracy and utility-based ranking metrics across various experimental settings. We also demonstrate that this performance improvement is not achieved at the expense of other popular performance measures, such as catalog coverage and aggregate diversity. The proposed approach can also be applied to other popular methods for rating estimation.
Improving user profile with personality traits predicted from social media content BIBAFull-Text 355-358
  Rui Gao; Bibo Hao; Shuotian Bai; Lin Li; Ang Li; Tingshao Zhu
Existing studies indicate that there exists strong correlation between personality and personal preference, thus personality could potentially be used to build more personalized recommender system. Personality traits are mainly measured by psychological questionnaires, and it is hard to obtain personality traits of large amount of users in real-world scenes.In this paper, we propose a new approach to automatically identify personality traits with Social Media contents in Chinese language environments. Social Media content features were extracted from 1766 Sina micro blog users, and the predicting model is trained with machine learning algorithms.The experimental results demonstrate that users' personality traits could be predicted from Social Media contents with acceptable Pearson Correlation, which makes it possible to develop user profiles for recommender system. In future, user profiles with predicted personality traits would be used to enhance the performance of existing personalized recommendation systems.
Leveraging the citation graph to recommend keywords BIBAFull-Text 359-362
  Ido Blank; Lior Rokach; Guy Shani
Users of scientific papers databases, such as CiteSeer, Google Scholar, and Microsoft Academic, often search for papers using a set of keywords. Unfortunately, many authors avoid listing sufficient keywords for their papers. As such, these applications may need to automatically associate good descriptive keywords with papers. This is a well-studied problem given the complete text of the paper, but in many cases, due to copyright privileges, research papers databases do not have the complete text, only metadata, such as the title and abstract. On the other hand, research papers databases typically maintain the citation network of each paper. In this paper we study the problem of predicting which keywords are appropriate for a scientific paper, using only the citation network. We compare our method with predicting keywords using the title and abstract, concluding that the citation network provides much better predictions.
Local context modeling with semantic pre-filtering BIBAFull-Text 363-366
  Victor Codina; Francesco Ricci; Luigi Ceccaroni
Context-Aware Recommender Systems locally adapt to a specific contextual situation the rating prediction computed by a traditional context-free recommender. In this paper we present a novel semantic pre-filtering approach that can be tuned to the optimal level of contextualization by aggregating contextual situations that are similar to the target one. The similarities of contextual situations are derived from the available contextually tagged users' ratings according to how similarly the contextual conditions influence the user's rating behavior. We present an extensive evaluation of the performance of our pre-filtering approach on several data sets, showing that it outperforms state-of-the-art context-aware Matrix Factorization approaches.
Musical recommendations and personalization in a social network BIBAFull-Text 367-370
  Dmitry Bugaychenko; Alexandr Dzuba
This paper presents a set of algorithms used for music recommendations and personalization in a general purpose social network www.ok.ru, the second largest social network in the CIS visited by more then 40 millions users per day. In addition to classical recommendation features like "recommend a sequence" and "find similar items" the paper describes novel algorithms for construction of context aware recommendations, personalization of the service, handling of the cold-start problem, and more. All algorithms described in the paper are working on-line and are able to detect and address changes in the user's behavior and needs in the real time.
   The core component of the algorithms is a taste graph containing information about different entities (users, tracks, artists, etc.) and relations between them (for example, user A likes song B with certainty X, track B created by artist C, artist C is similar to artist D with certainty Y and so on). Using the graph it is possible to select tracks a user would most probably like, to arrange them in a way that they match each other well, to estimate which items from a fixed list are most relevant for the user, and more.
   In addition, the paper describes the approach used to estimate algorithms efficiency and analyze the impact of different recommendation related features on the users' behavior and overall activity at the service.
Not by search alone: how recommendations complement search results BIBAFull-Text 371-374
  Daria Dzyabura; Alex Tuzhilin
This paper presents a novel approach to combining search and recommendations methods into one integrated system to satisfy user information seeking needs. It is shown theoretically and experimentally using simulations that the proposed combined approach outperforms "pure" search and "pure" recommendations in those cases when search is hindered by the user's inability to come up with a complete set of search criteria, and recommendation engine produces mediocre results.
OFF-set: one-pass factorization of feature sets for online recommendation in persistent cold start settings BIBAFull-Text 375-378
  Michal Aharon; Natalie Aizenberg; Edward Bortnikov; Ronny Lempel; Roi Adadi; Tomer Benyamini; Liron Levin; Ran Roth; Ohad Serfaty
One of the most challenging recommendation tasks is recommending to a new, previously unseen user. This is known as the user cold start problem. Assuming certain features or attributes of users are known, one approach for handling new users is to initially model them based on their features.
   Motivated by an ad targeting application, this paper describes an extreme online recommendation setting where the cold start problem is perpetual. Every user is encountered by the system just once, receives a recommendation, and either consumes or ignores it, registering a binary reward.
   We introduce One-pass Factorization of Feature Sets, 'OFF-Set', a novel recommendation algorithm based on Latent Factor analysis, which models users by mapping their features to a latent space. OFF-Set is able to model non-linear interactions between pairs of features, and updates its model per each recommendation-reward observation in a pure online fashion. We evaluate OFF-Set against several state of the art baselines, and demonstrate its superiority on real ad-targeting data.
Personalised ranking with diversity BIBAFull-Text 379-382
  Neil J. Hurley
In this paper we discuss a method to incorporate diversity into a personalised ranking objective, in the context of ranking-based recommendation using implicit feedback. The goal is to provide a ranking of items that respects user preferences while also tending to rank diverse items closely together. A prediction formula is learned as the product of user and item feature vectors, in order to minimise the mean squared error objective used previously in the RankALS and RankSGD methods, but modified to weight the difference in ratings between two items by the dissimilarity of those items. We report on preliminary experiments with this modified objective, in which the minimisation is carried out using stochastic gradient descent. We show that rankings based on the output of the minimisation succeed in producing recommendation lists with greater diversity, with just a small loss in relevance of the recommendation, as measured by the error rate.
Prior ratings: a new information source for recommender systems in e-commerce BIBAFull-Text 383-386
  Guibing Guo; Jie Zhang; Daniel Thalmann; Neil Yorke-Smith
Lack of motivation to provide ratings and eligibility to rate generally only after purchase restrain the effectiveness of recommender systems and contribute to the well-known data sparsity and cold start problems. This paper proposes a new information source for recommender systems, called prior ratings. Prior ratings are based on users' experiences of virtual products in a mediated environment, and they can be submitted prior to purchase. A conceptual model of prior ratings is proposed, integrating the environmental factor presence whose effects on product evaluation have not been studied previously. A user study conducted in website and virtual store modalities demonstrates the validity of the conceptual model, in that users are more willing and confident to provide prior ratings in virtual environments.
Probabilistic collaborative filtering with negative cross entropy BIBAFull-Text 387-390
  Alejandro Bellogin; Javier Parapar; Pablo Castells
Relevance-Based Language Models are an effective IR approach which explicitly introduces the concept of relevance in the statistical Language Modelling framework of Information Retrieval. These models have shown to achieve state-of-the-art retrieval performance in the pseudo relevance feedback task. In this paper we propose a novel adaptation of this language modeling approach to rating-based Collaborative Filtering. In a memory-based approach, we apply the model to the formation of user neighbourhoods, and the generation of recommendations based on such neighbourhoods. We report experimental results where our method outperforms other standard memory-based algorithms in terms of ranking precision.
Recommending improved configurations for complex objects with an application in travel planning BIBAFull-Text 391-394
  Amihai Savir; Ronen Brafman; Guy Shani
Users often configure complex objects with many possible internal choices. Recommendation engines that automatically configure such objects given user preferences and constraints, may provide much value in such cases. These applications generate appropriate recommendations based on user preferences. It is likely, though, that the user will not be able to fully express her preferences and constraints, requiring a phase of manual tuning of the recommended configuration. We suggest that following this manual revision, additional constraints and preferences can be automatically collected, and the recommended configuration can be automatically improved. Specifically, we suggest a recommender component that takes as input an initial manual configuration of a complex object, deduces certain user preferences and constraints from this configuration, and constructs an alternative configuration. We show an appealing application for our method in complex trip planning, and demonstrate its usability in a user study.
Recommending patents based on latent topics BIBAFull-Text 395-398
  Ralf Krestel; Padhraic Smyth
The availability of large volumes of granted patents and applications, all publicly available on the Web, enables the use of sophisticated text mining and information retrieval methods to facilitate access and analysis of patents. In this paper we investigate techniques to automatically recommend patents given a query patent. This task is critical for a variety of patent-related analysis problems such as finding relevant citations, research of relevant prior art, and infringement analysis. We investigate the use of latent Dirichlet allocation and Dirichlet multinomial regression to represent patent documents and to compute similarity scores. We compare our methods with state-of-the-art document representations and retrieval techniques and demonstrate the effectiveness of our approach on a collection of US patent publications.
Recommending scientific articles using bi-relational graph-based iterative RWR BIBAFull-Text 399-402
  Geng Tian; Liping Jing
The overabundance of scientific article information has created much inconvenience to researchers seeking interesting articles online. In this paper, we provide a Bi-Relational graph to represent the heterogenous information of scientific article recommendation system, which includes three parts: the article content similarity, researcher interest correlation, and researcher-article readership. Meanwhile, an iterative random walk with restarts learning method is proposed on the Bi-Relational graph to recommend a researcher rating for each article by making use of the known information. The proposed method has ability to perform both old and new article recommendation. A series of experiments on CiteULike dataset have shown that our method is more effective than other testing methods in the paper.
Sample selection for MCMC-based recommender systems BIBAFull-Text 403-406
  Thierry Silbermann; Immanuel Bayer; Steffen Rendle
Bayesian Inference with Markov Chain Monte Carlo (MCMC) has been shown to provide high prediction quality in recommender systems. The advantage over learning methods such as coordinate descent/alternating least-squares (ALS) or (stochastic) gradient descent (SGD) is that MCMC takes uncertainty into account and moreover MCMC can easily integrate priors to learn regularization values. For factorization models, MCMC inference can be done with efficient Gibbs samplers.
   However, MCMC algorithms are not point estimators, but they generate a chain of models. The whole chain of models is used to calculate predictions. For large scale models like factorization methods with millions or billions of model parameters, saving the whole chain of models is very storage intensive and can even get infeasible in practice. In this paper, we address this problem and show how a small subset from the chain of models can approximate the predictive distribution well. We use the fact that models from the chain are correlated and propose online selection techniques to store only a small subset of the models. We perform an empirical analysis on the large scale Netflix dataset with several Bayesian factorization models, including matrix factorization and SVD++. We show that the proposed selection techniques approximate the predictions well with only a small subset of model samples.
Selecting content-based features for collaborative filtering recommenders BIBAFull-Text 407-410
  Royi Ronen; Noam Koenigstein; Elad Ziklik; Nir Nice
We study the problem of scoring and selecting content-based features for a collaborative filtering (CF) recommender system. Content-based features play a central role in mitigating the "cold start" problem in commercial recommenders. They are also useful in other related tasks, such as recommendation explanation and visualization. However, traditional feature selection methods do not generalize well to recommender systems. As a result, commercial systems typically use manually crafted and selected features. This work presents a framework for automated selection of informative content-based features, that is independent of the type of recommender system or the type of features. We evaluate on recommenders from different domains: books, movies and smart-phone apps, and show effective results on each. In addition, we show how to use the proposed methods to generate meaningful features from text.
Sentimental product recommendation BIBAFull-Text 411-414
  Ruihai Dong; Michael P. O'Mahony; Markus Schaal; Kevin McCarthy; Barry Smyth
This paper describes a novel approach to product recommendation that is based on opinionated product descriptions that are automatically mined from user-generated product reviews. We present a recommendation ranking strategy that combines similarity and sentiment to suggest products that are similar but superior to a query product according to the opinion of reviewers. We demonstrate the benefits of this approach across a variety of Amazon product domains.
Set-oriented personalized ranking for diversified top-n recommendation BIBAFull-Text 415-418
  Ruilong Su; Li'Ang Yin; Kailong Chen; Yong Yu
In this paper, we propose a set-oriented personalized ranking model for diversified top-N recommendation. Users may have various individual ranges of interests. For personalized top-N recommendation task, the combination of relevance and diversity in recommendation results would be desirable. For this purpose, we integrate the concept of diversity into traditional matrix factorization model to construct a set-oriented collaborative filtering model. By optimizing this model with a set-oriented pairwise ranking method, we directly achieve personalized top-N recommendation results which are both relevant and diversified. We also utilize category information explicitly for learning personalized diversity. Experimental results show that our model outperforms traditional models in terms of personalized diversity and maintains good performance on relevance prediction.
Towards scalable and accurate item-oriented recommendations BIBAFull-Text 419-422
  Noam Koenigstein; Yehuda Koren
Most recommenders research aims at personalized systems, which suggest items based on user profiles. However, in reality many systems deal with item-oriented recommendations. In such setups, given a single item of interest, the system needs to provide other related items, following patterns like "people who liked this also liked...".
   While item-oriented systems are central in their importance, they have been approached so far using very basic tools. We identify several hurdles faced by standard approaches to the item-oriented task. First, the sparseness of observed activities prevents establishing reliable similarity relations for many item pairs. Second, we address a scalability challenge at the retrieval stage present in many real-world systems: Given an item inventory, which may encompass millions of items, it is desired to identify the most related item pairs in a sub-quadratic time. This work addresses these two challenges, thereby improving both accuracy and scalability of item-oriented recommenders. Additionally, we propose an empirical evaluation scheme for comparing the quality of different solutions with encouraging results.
Using geospatial metadata to boost collaborative filtering BIBAFull-Text 423-426
  Alexander Ostrikov; Lior Rokach; Bracha Shapira
In this paper, we present a method for boosting collaborative filtering by integrating spatial information about geo-referenced items (e.g., photos). In particular, we developed a method to estimate missing ratings by propagating an item's neighbor's ratings based on the similarity of geospatial information. An empirical evaluation shows that geospatial information significantly improves recommendation results, and its contribution grows with the ratings data's level of sparseness. We illustrate the usefulness of the method for a photo recommendation task using data obtained from two popular photo-sharing websites: Flickr and Panoramio. A comparison with state-of-the-art methods indicates the superiority of the proposed method, implying that geospatial information should be considered, when available.
When power users attack: assessing impacts in collaborative recommender systems BIBAFull-Text 427-430
  David C. Wilson; Carlos E. Seminario
Power users, in a Collaborative Filtering (CF) Recommender System (RS) context, are those who can exert considerable influence over the recommendation outcomes presented to other users. RS operators encourage the existence of power user communities and leverage them to help fellow users make informed purchase decisions. Thus, RS research in this area has focused on power user identification and utilization to address challenges such as rating sparsity for new items or users. But, as ever, there remains the potential for corruption of power. Alongside accuracy and efficiency measures, RS robustness to manipulation or 'attack' has been studied using injection of false user profiles. Our research is investigating the impact on RS predictions and top-N recommendation lists when simulated power users provide biased ratings for new items. In this study, we introduce the notion of a 'Power User Attack' for RS robustness analysis, as well as a novel use of social networking degree centrality concepts for identifying RS power users. Initial results show that power users identified using in-degree centrality, compared to other techniques, can be more influential as reflected by accuracy and robustness impacts before and after power user attacks.
xCLiMF: optimizing expected reciprocal rank for data with multiple levels of relevance BIBAFull-Text 431-434
  Yue Shi; Alexandros Karatzoglou; Linas Baltrunas; Martha Larson; Alan Hanjalic
Extended Collaborative Less-is-More Filtering xCLiMF is a learning to rank model for collaborative filtering that is specifically designed for use with data where information on the level of relevance of the recommendations exists, e.g. through ratings. xCLiMF can be seen as a generalization of the Collaborative Less-is-More Filtering (CLiMF) method that was proposed for top-N recommendations using binary relevance (implicit feedback) data. The key contribution of the xCLiMF algorithm is that it builds a recommendation model by optimizing Expected Reciprocal Rank, an evaluation metric that generalizes reciprocal rank in order to incorporate user feedback with multiple levels of relevance. Experimental results on real-world datasets show the effectiveness of xCLiMF, and also demonstrate its advantage over CLiMF when more than two levels of relevance exist in the data.
Evolving friend lists in social networks BIBAFull-Text 435-438
  Jacob W. Bartel; Prasun Dewan
In a social network, users can sort members of their social graph into friend lists to both understand the social structures within the graph and control the flow of incoming and outgoing information. To reduce the user-effort required to create these lists, previous work has developed techniques for generating friend-lists in a static social graph. This paper considers the user effort required to create friend lists in an evolving graph. We have developed several new initial quantitative metrics to capture this effort, and identified an initial technique for modeling graph growth. We have used these metrics and model to compare two techniques for evolving friend lists when the social graph grows: manual evolution -- the user evolves friend lists using no external tools -- and full recommendation -- an existing state of the art tool recommends a whole new set of friend lists. In these comparisons, we used the friend lists of 12 individuals, and simulated the growth of their social graphs and friend lists using our graph-growth model. Intuitively, when the graph evolves by a small (large) amount, the manual (automatic) approach should perform better. Our experiments show that full recommendation performs better than manual when the social graph changes by more than 1%, and yields an almost complete reduction in effort in the best cases.
Exploratory and interactive daily deals recommendation BIBAFull-Text 439-442
  Anisio Lacerda; Adriano Veloso; Nivio Ziviani
Daily deals sites (DDSs), such as Groupon and LivingSocial, attract millions of customers in the hunt for products and services at significantly reduced prices. A typical approach to increase revenue is to send email messages featuring the deals of the day. Such daily messages, however, are usually not centered on the customers, instead, all registered users typically receive similar messages with almost the same deals. Traditional recommendation algorithms are innocuous in DDSs because: (i) most of the users are sporadic bargain hunters, and thus past preference data is extremely sparse, (ii) deals have a short living period, and thus data is extremely volatile, and (iii) user taste and interest may undergo temporal drifts. In order to address such particularly challenging scenario, we propose new algorithms for daily deals recommendation based on the explore-then-exploit strategy.Users are split into exploration and exploitation sets -- in the exploration set the users receive non-personalized messages and a co-purchase network is updated with user feedback for purchases of the day, while in the exploitation set the updated network is used for recommending personalized messages for the remaining users.A thorough evaluation of our algorithms using real data obtained from a large daily deals website in Brazil in contrast to state-of-the-art recommendation algorithms show gains in precision ranging from 18% to 34%.

Doctoral symposium

Dynamic generation of personalized hybrid recommender systems BIBAFull-Text 443-446
  Simon Dooms
The problem of information overload has been a relevant and active research topic for the past twenty years. Since then, numerous algorithms and recommendation approaches have been proposed, which gives rise to a new type of problem: recommendation algorithm overload. Although hybrid recommendation techniques, which combine the strengths of individual recommenders, have become well-accepted, the procedure of building and tuning a hybrid recommender is still a tedious and time-consuming process. In our work, we focus on dynamically building personalized hybrid recommender systems on an individual user basis. By means of a dynamic online learning strategy we combine the most appropriate recommendation algorithms for a user based on realtime relevance feedback. Learning effectiveness of genetic algorithms, machine learning techniques and other optimization approaches will be studied in both an offline and online setting.
Accuracy and robustness impacts of power user attacks on collaborative recommender systems BIBAFull-Text 447-450
  Carlos E. Seminario
Power users, in a Collaborative Filtering (CF) Recommender System (RS) context, are those who can exert considerable influence over the recommendation outcomes presented to other users. RS operators encourage the existence of power user communities and leverage them to help fellow users make informed purchase decisions. Thus, RS research in this area has focused on power user selection and utilization to address challenges such as rating sparsity for new items or users. But, as ever, there remains the potential for corruption by power users who provide biased ratings. And because of the influence that power users wield, biased ratings they provide can have significant impacts on RS accuracy and robustness. In order to better understand this problem and develop solution strategies, our research is investigating the impact on RS predictions and top-N recommendation lists when power users provide biased ratings. The open areas of research we have explored are analyzing and evaluating power user selection techniques, statistically characterizing power users in order to create attack profiles, mounting power user attacks on new items, and using accuracy and robustness metrics to evaluate power user attacks. In the future, we plan to extend our initial research in power user selection, characterization, and evaluation, as well as generate attack profiles based on power user characteristics, mount power user attacks on user-based, item-based, and SVD-based CF systems, evaluate power user attacks, and generalize our work across different domains.
Integrating trust and similarity to ameliorate the data sparsity and cold start for recommender systems BIBAFull-Text 451-454
  Guibing Guo
Our research aims to tackle the problems of data sparsity and cold start of traditional recommender systems. Insufficient ratings often result in poor quality of recommendations in terms of accuracy and coverage. To address these issues, we propose three different approaches from the perspective of preference modelling. Firstly, we propose to merge the ratings of trusted neighbors and thus form a new rating profile for the active users, based on which better recommendations can be generated. Secondly, we aim to make better use of user ratings and introduce a novel Bayesian similarity measure by taking into account both the direction and length of rating vectors. Thirdly, we propose a new information source called prior ratings based on virtual product experience in virtual reality environments, in order to inherently resolve the concerned problems.
Agent-based computational investing recommender system BIBAFull-Text 455-458
  Mona Taghavi; Kaveh Bakhtiyari; Edgar Scavino
The fast development of computing and communication has reformed the financial markets' dynamics. Nowadays many people are investing and trading stocks through online channels and having access to real-time market information efficiently. There are more opportunities to lose or make money with all the stocks information available throughout the World; however, one should spend a lot of effort and time to follow those stocks and the available instant information. This paper presents a preliminary regarding a multi-agent recommender system for computational investing. This system utilizes a hybrid filtering technique to adaptively recommend the most profitable stocks at the right time according to investor's personal favour. The hybrid technique includes collaborative and content-based filtering. The content-based model uses investor preferences, influencing macro-economic factors, stocks profiles and the predicted trend to tailor to its advices. The collaborative filter assesses the investor pairs' investing behaviours and actions that are proficient in economic market to recommend the similar ones to the target investor.
Beyond rating prediction accuracy: on new perspectives in recommender systems BIBAFull-Text 459-462
  Panagiotis Adamopoulos
This paper proposes a number of studies in order to move recommender systems beyond the traditional paradigm and the classical perspective of rating prediction accuracy. We contribute to existing helpful but less explored paradigms and also propose new approaches aiming at more useful recommendations for both users and businesses. Working toward this direction, we discuss the studies we have conducted so far and present our future research plans. In particular, we move our focus from even more accurate rating predictions and aim at offering a holistic experience to the users by avoiding the over-specialization of generated recommendations and providing the users with sets of non-obvious but high quality recommendations that fairly match their interests and they will remarkably like.
Anytime algorithms for top-N recommenders BIBAFull-Text 463-466
  David Ben-Shimon
Many small and mid-sized e-businesses use the services of recommender system (RS) provider companies to outsource the construction and maintenance of their RS. The fees that RS providers charge their clients must cover the computation costs for constructing and updating the recommendation model. By using anytime algorithms, a RS provider can control the computation costs and still offer a system capable of delivering reasonable recommendations. Thus, a RS provider should be able to stop the construction of a recommendation model once the cost for computing it reaches the amount the customer has agreed to pay. In this research we suggest anytime algorithms as a possible solution to a problem that RS providers face. We demonstrate how certain existing recommendation algorithms can be adjusted to the anytime framework. We focus on the case of item-item algorithms, showing how the anytime behavior can be improved using different ordering methods of computations. We conduct a comparative study demonstrating the benefits of the proposed methods for top-N item-item recommenders.


GAIN: web service for user tracking and preference learning -- a smart TV use case BIBAFull-Text 467-468
  Jaroslav Kuchar; Tomáš Kliegr
GAIN (inbeat.eu) is a web application and service for capturing and preprocessing user interactions with semantically described content. GAIN outputs a set of instances in tabular form suitable for further processing with generic machine-learning algorithms. GAIN is demoed as a component of a "SMART-TV" recommender system. Content is automatically described with DBpedia types using a Named Entity Recognition (NER) system. Interest is determined based on explicit user actions and user's attention computed by 3D head pose estimation. Preference rules are learnt with an association rule mining algorithm. These can be e.g. deployed to a business rules system, acting as a recommender.
PEN RecSys: a personalized news recommender systems framework BIBAFull-Text 469-470
  Florent Garcin; Boi Faltings
We present the Personalized News (PEN) recommender systems framework, currently in use by a newspaper website to evaluate various algorithms for news recommendations. We briefly describe its system architecture and related components. We show how a researcher can easily evaluate different algorithms thanks to a web-based interface.
A heterogeneous graph-based recommendation simulator BIBAFull-Text 471-472
  Yeonchan Ahn Ahn; Sungchan Park; Sangkeun Lee; Sang-goo Lee
Heterogeneous graph-based recommendation frameworks have flexibility in that they can incorporate various recommendation algorithms and various kinds of information to produce better results. In this demonstration, we present a heterogeneous graph-based recommendation simulator which enables participants to experience the flexibility of a heterogeneous graph-based recommendation method. With our system, participants can simulate various recommendation semantics by expressing the semantics via meaningful paths like User → Movie → User → Movie. The simulator then returns the recommendation results on the fly based on the user-customized semantics using a fast Monte Carlo algorithm.
Design and evaluation of a client-side recommender system BIBAFull-Text 473-474
  Chris Newell; Libby Miller
Most recommender systems found on the web are server-based and centralised. However, it can be difficult to maintain the responsiveness with this approach when there are large numbers of concurrent users. In this demonstration we present an alternative approach where major parts of the recommender system are implemented in scripts run by the user's client system.
Sage: recommender engine as a cloud service BIBAFull-Text 475-476
  Royi Ronen; Noam Koenigstein; Elad Ziklik; Mikael Sitruk; Ronen Yaari; Neta Haiby-Weiss
Project Sage is Microsoft's all-purpose recommender system, designed and deployed as an ultra-high scale cloud service. Sage focuses on both state of the art research and high scale robust implementation. In the research front, we demonstrate new pre-processing and cleaning techniques, a novel probabilistic matrix factorization model for implicit one-class data, and a relatively new evaluation framework. In the engineering front, we present a working service deployed on the Microsoft Azure cloud, which provides easy-to-use interfaces to integrate a recommendation service into any website.


The fifth ACM RecSys workshop on recommender systems and the social web BIBFull-Text 477-478
  Bamshad Mobasher; Dietmar Jannach; Werner Geyer; Jill Freyne; Andreas Hotho; Sarabjot Singh Anand; Ido Guy
Workshop on human decision making in recommender systems: decisions@RecSys'13 BIBAFull-Text 479-480
  Li Chen; Marco de Gemmis; Alexander Felfernig; Pasquale Lops; Francesco Ricci; Giovanni Semeraro; Martijn C. Willemsen
A primary function of recommender systems is to help their users to make better choices and decisions. The overall goal of the workshop is to analyse and discuss novel techniques and approaches for supporting effective and efficient human decision making in different types of recommendation scenarios. The submitted papers discuss a wide range of topics from core algorithmic issues to the management of the human computer interaction.
Workshop and challenge on news recommender systems BIBAFull-Text 481-482
  Mozhgan Tavakolifard; Jon Atle Gulla; Kevin C. Almeroth; Frank Hopfgartner; Benjamin Kille; Till Plumbaum; Andreas Lommatzsch; Torben Brodt; Arthur Bucko; Tobias Heintz
Recommending news articles entails additional requirements to recommender systems. Such requirements include special consumption patterns, fluctuating item-collections, and highly sparse user profiles. This workshop (NRS'13@RecSys) brought together researchers and practitioners around the topics of designing and evaluating novel news recommender systems. Additionally, we offered a challenge allowing participants to evaluate their recommendation algorithms with actual user feedback.
Workshop on recommender systems meet big data & semantic technologies: SeRSy 2013 BIBAFull-Text 483-484
  Marco de Gemmis; Tommaso Di Noia; Ora Lassila; Pasquale Lops; Thomas Lukasiewicz; Giovanni Semeraro
The primary goal of the workshop is to showcase cutting edge research on the intersection of Recommender Systems and Semantic Technologies, by taking the best of the two worlds. This combination may provide the RecSys community with important scenarios where the potential of Semantic Technologies can be effectively exploited into systems performing complex tasks, such as recommendation engines processing Big Data.
Workshop on reproducibility and replication in recommender systems evaluation: RepSys BIBAFull-Text 485-486
  Alejandro Bellogin; Pablo Castells; Alan Said; Domonkos Tikk
Experiment replication and reproduction are key requirements for empirical research methodology, and an important open issue in the field of Recommender Systems. When an experiment is repeated by a different researcher and exactly the same result is obtained, we can say the experiment has been replicated. When the results are not exactly the same but the conclusions are compatible with the prior ones, we have a reproduction of the experiment. Reproducibility and replication involve recommendation algorithm implementations, experimental protocols, and evaluation metrics. While the problem of reproducibility and replication has been recognized in the Recommender Systems community, the need for a clear solution remains largely unmet, which motivates the present workshop.
First workshop on large-scale recommender systems: research and best practice (LSRS 2013) BIBAFull-Text 487-488
  Tao Ye; Danny Bickson; Quan Yuan
With the increase of data collected and computation power available, modern recommender systems are ever facing new challenges. While complex models are developed in academia, industry practice seems to focus on relatively simple techniques that can deal with the magnitude of data and the need to distribute the computation. The first workshop on large-scale recommender systems (LSRS) is a meeting place for industry and academia to discuss the current and future challenges of applied large-scale recommender systems.
RecSys challenge 2013 BIBFull-Text 489-490
  Jim Blomo; Martin Ester; Marty Field


Recommendation in social networks BIBFull-Text 491-492
  Martin Ester
Learning to rank for recommender systems BIBAFull-Text 493-494
  Alexandros Karatzoglou; Linas Baltrunas; Yue Shi
Recommender system aim at providing a personalized list of items ranked according to the preferences of the user, as such ranking methods are at the core of many recommendation algorithms. The topic of this tutorial focuses on the cutting-edge algorithmic development in the area of recommender systems. This tutorial will provide an in depth picture of the progress of ranking models in the field, summarizing the strengths and weaknesses of existing methods, and discussing open issues that could be promising for future research in the community. A qualitative and quantitative comparison between different models will be provided while we will also highlight recent developments in the areas of Reinforcement Learning.
Beyond friendship: the art, science and applications of recommending people to people in social networks BIBAFull-Text 495-496
  Luiz Augusto Pizzato; Anmol Bhasin
While Recommender Systems are powerful drivers of engagement and transactional utility in social networks, People recommenders are a fairly involved and diverse subdomain. Consider that movies are recommended to be watched, news is recommended to be read, people however, are recommended for a plethora of reasons -- such as recommendation of people to befriend, follow, partner, targets for an advertisement or service, recruiting, partnering romantically and to join thematic interest groups.
   This tutorial aims to first describe the problem domain, touch upon classical approaches like link analysis and collaborative filtering and then take a rapid deep dive into the unique aspects of this problem space like reciprocity, intent understanding of recommender and the recomendee, contextual people recommendations in communication flows and social referrals -- a paradigm for delivery of recommendations using the social graph. These aspects will be discussed in the context of published original work developed by the authors and their collaborators and in many cases deployed in massive-scale real world applications on professional networks such as LinkedIn.
Tutorial on preference handling BIBAFull-Text 497-498
  Alexis Tsoukiàs; Paolo Viappiani
This tutorial, addressed to researchers and practitioners in the area of recommendation systems, reviews the foundations of decision and utility theory, focusing on methods for representing, learning and reasoning with preferences. Recent advances on handling preferences will also be discussed, with a focus on preference learning, preference aggregation and preferences in argumentation. Finally, we will show how these techniques can be used in a practice in recommender systems.