HCI Bibliography Home | HCI Conferences | SocInfo Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
SocInfo Tables of Contents: 1011121313w14

Proceedings of the 2012 International Conference on Social Informatics

Fullname:SocInfo 2012: 4th International Conference on Social Informatics
Editors:Karl Aberer; Andreas Flache; Wander Jager; Ling Liu; Jie Tang; Christophe Guéret
Location:Lausanne, Switzerland
Dates:2012-Dec-05 to 2012-Dec-07
Publisher:Springer Berlin Heidelberg
Series:Lecture Notes in Computer Science 7710
Standard No:DOI: 10.1007/978-3-642-35386-4 hcibib: SocInfo12; ISBN: 978-3-642-35385-7 (print), 978-3-642-35386-4 (online)
Links:Online Proceedings | Conference Website
A System for Web Widget Discovery Using Semantic Distance between User Intent and Social Tags BIBAKFull-Text 1-14
  Zhenzhen Zhao; Xiaodi Huang; Noël Crespi
Social interaction leverages collective intelligence through user-generated content, social networking, and social annotation. Users are enabled to enrich knowledge representation by rating, commenting, and tagging. The existing systems for service discovery make use of semantic relation among social tags, but ignore the relation between a user information need for services and tags. This paper first provides an overview of how social tagging is applied to discover contents/services. An enhanced web widget discovery model that aims to discover services mostly relevant to users is then proposed. The model includes an algorithm that quantifies the accurate relation between user intent for a service and the tags of a widget, as well as three different widget discovery schemes. Using the online service of Widgetbox.com, we experimentally demonstrate the accuracy and efficiency of our system.
Keywords: content discovery; folksonomy; service discovery; social tagging; algorithm; widget
An Automated Multiscale Map of Conversations: Mothers and Matters BIBAKFull-Text 15-28
  Ansuya Ahluwalia; Allen Huang; Roja Bandari; Vwani Roychowdhury
By augmenting conventional techniques of topic modeling with unigram analysis and community detection, we establish an automated method that generates a comprehensive and meaningful summary of forum conversations over time that also sheds light on patterns of user behavior. We combine these methods to obtain a multiscale representation of what topics are being discussed, what the users are saying about each topic, how the conversation is evolving over time, and how friendships relate to content. As an example of our methodology, we examine discussion boards on Cafemom -- an online hub for women to share their experiences and discuss their views on issues pertinent to child rearing. We apply the method with a focus on the issue of vaccination -- a subject matter which has become controversial in recent years. We demonstrate how our methodology provides valuable insights into the evolution of conversations and highlights similarities in attitudes of socially connected users.
Keywords: Topic Modeling; Community Detection; Content Analysis; Vaccination; Forums
How Influential Are You: Detecting Influential Bloggers in a Blogging Community BIBAKFull-Text 29-42
  Imrul Kayes; Xiaoning Qian; John Skvoretz; Adriana Iamnitchi
Blogging is a popular activity with high impact on marketing, shaping public opinions, and informing the world about major events from a grassroots point of view. Influential bloggers are recognized by businesses as significant forces for product promotion or demotion, and by oppressive political regimes as serious threats to their power. This paper studies the problem of identifying influential bloggers in a blogging community, BlogCatalog, by using network centrality metrics. Our analysis shows that bloggers are connected in a core-periphery network structure, with the highly influential bloggers well connected with each others forming the core, and the non-influential bloggers at the periphery. The six node centrality metrics we analyzed are highly correlated, showing that an aggregate centrality score as a measure of influence will be stable to variations in centrality metrics.
Keywords: social networks; influence; network centrality; blogosphere
A Simulation Model Using Transaction Cost Economics to Analyze the Impact of Social Media on Online Shopping BIBAKFull-Text 43-53
  Apratim Mukherjee; Shrabastee Banerjee; Somprakash Bandyopadhyay
In this paper, we have developed an agent-based simulation model to study the influence of social media on consumers' inclination towards on-line shopping. Social media includes web-based and mobile based technologies which are used to turn communication into interactive dialogue between organizations, communities, and individuals. Building upon the Transaction Cost Economics theory, the objective of our study is to examine the effect of social media on the "perceived transaction cost" of an individual, which determines his/her inclination to buy online. Transaction cost economics (TCE) theoretically explains why a transaction subject favors a particular form of transaction over others. Since purchasing from online stores can be considered a choice between the internet and traditional stores, it is reasonable to assume that consumers will go with the channel that has the lower transaction cost. Using agent-based models, we have studied the rate of adoption of on-line shopping by consumers and found it to be exponential, not linear.
Keywords: Transaction Cost; Social Network; Online shopping; Consumer Behavior
Predicting Group Evolution in the Social Network BIBAKFull-Text 54-67
  Piotr Bródka; Przemyslaw Kazienko; Bartosz Koloszczyk
Groups -- social communities are important components of entire societies, analysed by means of the social network concept. Their immanent feature is continuous evolution over time. If we know how groups in the social network has evolved we can use this information and try to predict the next step in the given group evolution. In the paper, a new approach for group evolution prediction is presented and examined. Experimental studies on four evolving social networks revealed that (i) the prediction based on the simple input features may be very accurate, (ii) some classifiers are more precise than the others and (iii) parameters of the group evolution extracion method significantly influence the prediction quality.
Keywords: social network; group evolution; predicting group evolution; group dynamics; social network analysis; GED
Interpolating between Random Walks and Shortest Paths: A Path Functional Approach BIBAFull-Text 68-81
  François Bavaud; Guillaume Guex
General models of network navigation must contain a deterministic or drift component, encouraging the agent to follow routes of least cost, as well as a random or diffusive component, enabling free wandering. This paper proposes a thermodynamic formalism involving two path functionals, namely an energy functional governing the drift and an entropy functional governing the diffusion. A freely adjustable parameter, the temperature, arbitrates between the conflicting objectives of minimising travel costs and maximising spatial exploration. The theory is illustrated on various graphs and various temperatures. The resulting optimal paths, together with presumably new associated edges and nodes centrality indices, are analytically and numerically investigated.
Dynamic Targeting in an Online Social Medium BIBAFull-Text 82-95
  Peter Laflin; Alexander V. Mantzaris; Fiona Ainley; Amanda Otley; Peter Grindrod; Desmond J. Higham
Online human interactions take place within a dynamic hierarchy, where social influence is determined by qualities such as status, eloquence, trustworthiness, authority and persuasiveness. In this work, we consider topic-based Twitter interaction networks, and address the task of identifying influential players. Our motivation is the strong desire of many commercial entities to increase their social media presence by engaging positively with pivotal bloggers and Tweeters. After discussing some of the issues involved in extracting useful interaction data from a Twitter feed, we define the concept of an active node subnetwork sequence. This provides a time-dependent, topic-based, summary of relevant Twitter activity. For these types of transient interactions, it has been argued that the flow of information, and hence the influence of a node, is highly dependent on the timing of the links. Some nodes with relatively small bandwidth may turn out to be key players because of their prescience and their ability to instigate follow-on network activity. To simulate a commercial application, we build an active node subnetwork sequence based on key words in the area of travel and holidays. We then compare a range of network centrality measures, including a recently proposed version that accounts for the arrow of time, with respect to their ability to rank important nodes in this dynamic setting. The centrality rankings use only connectivity information (who Tweeted whom, when), but if we post-process the results by examining account details, we find that the time-respecting, dynamic, approach, which looks at the follow-on flow of information, is less likely to be 'misled' by accounts that appear to generate large numbers of automatic Tweets with the aim of pushing out web links. We then benchmark these algorithmically derived rankings against independent feedback from five social media experts who judge Twitter accounts as part of their professional duties. We find that the dynamic centrality measures add value to the expert view, and indeed can be hard to distinguish from an expert in terms of who they place in the top ten. We also highlight areas where the algorithmic approach can be refined and improved.
Connecting with Active People Matters: The Influence of an Online Community on Physical Activity Behavior BIBAKFull-Text 96-109
  Maartje Groenewegen; Dimo Stoyanov; Dirk Deichmann; Aart van Halteren
This paper discusses the impact of online social networks as means to motivate people to become more physically active. Based on a data set from 4333 participants we show that the activity level of people that participated in the online community (for 14 weeks) is significantly higher compared to people that choose not to become a member of that community. Detailed analyses show that the number of contacts in the online community does not have a significant effect on the physical activity level while network density even has a significant, negative effect. On the other hand, the activity level of a participant is higher when his or her friends also have a high average activity level. This effect is even higher when a participant's amount of friends increases. Theoretical and managerial implications concerning the impact of online social networks on offline behavior are discussed.
Keywords: Physical activity promotion; social network analyses; motivation
Note: Best paper award
Detecting Overlapping Communities in Location-Based Social Networks BIBAKFull-Text 110-123
  Zhu Wang; Daqing Zhang; Dingqi Yang; Zhiyong Yu; Xingshe Zhou
With the recent surge of location-based social networks (LBSNs, e.g., Foursquare, Facebook Places), huge amount of digital footprints about users' locations, profiles as well as their online social connections become accessible to service providers. Different from social networks (e.g., Flickr, Facebook) which have explicit groups for users to subscribe or join, LBSNs usually have no explicit community structure. In order to capitalize on the large number of potential users, quality community detection approach is needed so as to enable applications such as direct marketing, group tracking, etc. The diversity of people's interests and behaviors when using LBSNs suggests that their community structures overlap. In this paper, based on the user-venue check-in relationship and user/venue attributes, we come out with a novel multi-mode multi-attribute edge-centric co-clustering (M 2 Clustering) framework to discover the overlapping communities of LBSNs users. By employing inter-mode/intra-mode features, the proposed framework is able to group like-minded users from different social perspectives. The efficacy of our approach is validated by intensive empirical evaluations using the collected Foursquare dataset of 266,838 users with 9,803,764 check-ins over 2,477,122 venues worldwide.
Keywords: Community Detection; Overlapping Community; Edge-Clustering; Location-Based Social Networks
CrowdLang: A Programming Language for the Systematic Exploration of Human Computation Systems BIBAKFull-Text 124-137
  Patrick Minder; Abraham Bernstein
Human computation systems are often the result of extensive lengthy trial-and-error refinements. What we lack is an approach to systematically engineer solutions based on past successful patterns.
   In this paper we present the CrowdLang programming framework for engineering complex computation systems incorporating large crowds of networked humans and machines with a library of known interaction patterns. We evaluate CrowdLang by programming a German-to-English translation program incorporating machine translation and a monolingual crowd. The evaluation shows that CrowdLang is able to simply explore a large design space of possible problem-solving programs with the simple variation of the used abstractions. In an experiment involving 1918 different human actors, we show that the resulting translation program significantly outperforms a pure machine translation in terms of adequacy and fluency whilst translating more than 30 pages per hour and approximates the human-translated gold standard to 75%.
Keywords: CrowdLang; Programming Language; Human Computation; Collective Intelligence; Crowdsourcing; Translation Software
Experiments in Cross-Lingual Sentiment Analysis in Discussion Forums BIBAKFull-Text 138-151
  Hatem Ghorbel
One of the objectives of sentiment analysis is to classify the polarity of conveyed opinions from the perspective of textual evidence. Most of the work in the field has been intensively applied to the English language and only few experiments have explored other languages. In this paper, we present a supervised classification of posts in French online forums where sentiment analysis is based on shallow linguistic features such as POS tagging, chunking and common negation forms. Furthermore, we incorporate word semantic orientation extracted from the English lexical resource SentiWordNet as an additional feature. Since SentiWordNet is an English resource, lexical entries in the studied French corpus should be translated into English. For this purpose, we propose a number of French to English translation experiments such as machine translation and WordNet synset translation using EuroWordNet. Obtained results show that WordNet synset translation have not significantly improved the classification performance with respect to the bag of words baseline due to the shortage in coverage. Automatic translation haven't either significantly improved the results due to its insufficient quality. Propositions of improving the classification performance are given by the end of the article.
Keywords: Cross-Lingual Sentiment Analysis; Machine Translation; Supervised Classification
Quality Assessment of User Comments on Mobile Platforms Considering Channel of Activation and Platform Design BIBAKFull-Text 152-161
  Christopher Fröch; Martin Schumann
In this paper we present the results of an experimental three-step study concerning quality assessment of product reviews. As reviews and comments on products and services are gaining importance in the context of purchasing decisions providers of review platforms are seeking for ways to improve the quality of their platforms. In the presented study user expectations regarding quality of comments were collected as well as reader perceptions of quality. Additionally a thorough text analysis of experimentally obtained product reviews was conducted. The main results of this research were that quality expectations do not necessarily lead towards good quality comments provided by the same person. Moreover it could be observed that the combination of text and star rating is preferred by people and also will lead to better understandability of resulting comments. The channel of activation, NFC or QR codes did not cause any significant difference considering comment quality or appropriate platform.
Keywords: product reviews; quality assessment; content analysis; experiment
A Method Based on Congestion Game Theory for Determining Electoral Tendencies BIBAKFull-Text 162-173
  Guillermo De Ita; Luis Altamirano; Aurelio López-López; Yolanda Moyao
We present a novel method to study the tendencies of vote in sectorial democratic elections. Our method is intended to determine the relevant profiles characterizing the political behavior of voters. Those profiles allow us to model how the voters, in a specific election organized by sectors, make their vote decision. Furthermore, the same set of profiles are used for representing the different strategies applied by the candidates that compete in the election.
   We apply congestion games theory to simulate the distribution of the votes among the candidates, describing an automated way to estimate the likely number of votes for each candidate. Therefore, we can determine who will be the winner candidate of the election, according to a specific political scenario. We report the application of our model to simulate the elections of a director in a university setting, obtaining estimations very close to the actual outcomes.
Keywords: Social Behavior Modeling; Social Simulation; Electoral Simulation; Congestion Games; Multi-Agent System
A Model to Represent Human Social Relationships in Social Network Graphs BIBAKFull-Text 174-187
  Marco Conti; Andrea Passarella; Fabio Pezzoni
Human social relationships are a key component of emerging complex techno-social systems such as socially-centric platforms based on the interactions between humans and ICT technologies. Therefore, the models of human social relationships are fundamental to characterise these systems and study the performance of socially-centric platforms depending on the social context where they operate. The goal of this paper is presenting a generative model for building synthetic human social network graphs where the properties of social relationships are accurately reproduced. The model goes well beyond a binary approach, whereby edges between nodes, if existing, are all of the same type. It sets the properties of each social link, by incorporating fundamental results from the anthropology literature. The synthetic networks it generates accurately reproduce both the macroscopic structure (e.g., its diameter and clustering coefficient), and the microscopic structure (e.g., the properties of the tie strength of individual social links) of human social networks. We compare generated networks with a large-scale social network data set, validating that the model is able to produce graphs with the same structural properties of human-social-network graphs. Moreover, we characterise the impact of the model parameters on the synthetic graph properties.
Keywords: social networks; human behaviour; modelling; simulations
C4PS -- Helping Facebookers Manage Their Privacy Settings BIBAFull-Text 188-201
  Thomas Paul; Martin Stopczynski; Daniel Puscher; Melanie Volkamer; Thorsten Strufe
The ever increasing popularity of Online Social Networks has left a wealth of personal data on the web, accessible for broad and automatic retrieval. Protection from undesired recipients and harvesting by crawlers is implemented by access control, manually configured by the user in his privacy settings. Privacy unfriendly default settings and the user unfriendly privacy setting interfaces cause an unnoticed over-sharing. We propose C4PS -- Colors for Privacy Settings, a concept for future privacy setting interfaces. We developed a mockup for privacy settings in Facebook as a proof of concept, applying color coding for different privacy visibilities, providing easy access to the privacy settings, and generally following common, well known practices. We evaluated this mockup in a lab study and show in the results that the new approach increases the usability significantly. Based on the results we provide a Firefox plug-in implementing C4PS for the new Facebook interface.
Dynamic "Participative Rules" in Serious Games, New Ways for Evaluation? BIBAKFull-Text 202-215
  Jean-Pierre Cahier; Nour El Mawas; Aurélien Bénel
Rules are used by Computer Games to evaluate losses, gains, changing items and actions of the players. In addition, they reinforce realism and playability, especially in training situations where Knowledge is complex and expert (e.g. best practices acquisition in crisis management, decision making in complex socio-technical systems...). To evaluate items and actions, we propose a dynamic solution using "participative rules". In this approach, based on Computer Supported Cooperative Work and Knowledge Engineering, the rules base is directly generated from a special discussion forum which contains successive versions of the textual rules continuously discussed and co-built by the designers' community, in strong relation with the players' community. This paper resumes a "Work in progress" recently presented with more details [1] to the Game Community, but it extends it by adding the point that, beyond the "Serious Games" field, the notion of "participative rule" that we are exploring, could interest more broadly Human and Social Scientists who seek new ways towards effective evaluation methods.
Keywords: evaluation; participative rules; serious game; participative design
Mobile Phones, Family and Personal Relationships: The Case of Indonesian Micro-entrepreneurs BIBAKFull-Text 216-231
  Misita Anwar; Graeme Johanson
In the Indonesian context, the use of mobile phones has had many effects on the economic as well as the social fabric of communities. However, these impacts have not been thoroughly examined, particularly in relation to micro-enterprises, productivity or wellbeing. This paper evaluates the impact of mobiles from the perspective of human development where 'development' is seen as the expansion of people's choices and information and communications technologies (ICTs) are seen as supporting these choices. To test theories of development, it presents an empirical study undertaken in Indonesia about the impact of mobile phones on micro-entrepreneurs' wellbeing. Results show that micro-entrepreneurs regarded family is the most important aspect of their lives and that their own wellbeing was treated the same at that of their families. Accordingly, mobile phones are considered as a very significant force to maintain and improve their relationships with family, relatives and friends. Mobile phones contribute significantly to wellbeing.
Keywords: mobile phones; micro-entrepreneurs; human development; capability approach
An Analysis of Topical Proximity in the Twitter Social Graph BIBAKFull-Text 232-245
  Markus Schaal; John O'Donovan; Barry Smyth
Standard approaches of information retrieval are increasingly complemented by social search even when it comes to rational information needs. Twitter, as a popular source of real-time information, plays an important role in this respect, as both the follower-followee graph and the many relationships among users provide a rich set of information pieces about the social network. However, many hidden factors must be considered if social data are to successfully support the search for high-quality information. Here we focus on one of these factors, namely the relationship between content similarity and social distance in the social network. We compared two methods to compute content similarity among twitter users in a one-per-user document collection, one based on standard term frequency vectors, the other based on topic associations obtained by Latent Dirichlet Allocation (LDA). By comparing these metrics at different hop distances in the social graph we investigated the utility of prominent features such as Retweets and Hashtags as predictors of similarity, and demonstrated the advantages of topical proximity vs. textual similarity for friend recommendations.
Keywords: micro blogs; topical proximity; social network distance; friend recommendation
A Foresight Support System to Manage Knowledge on Information Society Evolution BIBAKFull-Text 246-259
  Andrzej M. J. Skulimowski
In this paper we present an intelligent knowledge fusion and decision support system tailored to manage information on future social and technological trends. It focuses on gathering and managing the rules that govern the evolution of selected information society technologies (IST) and their applications. The main idea of information gathering and processing here presented refers to so-called real-time expert Delphi, where an expert community works on the same research problems by responding to structured questionnaires, elaborating complex dynamical system models, providing recommendations, and verifying the models so arisen. The knowledge base is structured in layers that correspond to the selected kinds of information on the technology and social evolution, uses, markets, and management. An analytical engine uses labelled hypermultigraphs to process the mutual impacts of objects from each layer to elicit the technological evolution rules and calculate future trends and scenarios. The processing rules are represented within discrete-time and discrete-event control models. Multicriteria decision support procedures make it possible to aggregate individual expert recommendations. The resulting foresight support system can process uncertain information using a fuzzy-random-variable-based model, while a coupled reputation management system can verify collective expert judgments and assign trust vectors to experts and other sources of information.
Keywords: Foresight Support Systems; Complex Socioeconomic Models; Group Model Building; Knowledge Fusion; Intelligent Decision Support
How Many Answers Are Enough? Optimal Number of Answers for Q&A Sites BIBAKFull-Text 260-274
  Pnina Fichman
With the proliferation of the social web questions about information quality and optimization attract the attention of IS scholars. Question-answering (QA) sites, such as Yahoo!Answers, have the potential to produce good answers, but at the same time not all answers are good and not all QA sites are alike. When organizations design and plan for the integration of question answering services on their sites, identification of good answers and process optimization become critical. Arguing that 'given enough answers all questions are answered successfully,' this paper identifies the optimal number of posts that generate high quality answers. Based on content analysis of Yahoo! Answers' informational questions (n=174) and their answers (n=1,023), the study found that seven answers per question are 'enough' to provide a good answer.
Keywords: Q&A sites; CQA; Optimization; Web 2.0; Information Quality
Analysis and Support of Lifestyle via Emotions Using Social Media BIBAKFull-Text 275-291
  Ward van Breda; Jan Treur; Arlette van Wissen
Using recent insights from Cognitive, Affective and Social Neuroscience, this paper addresses how affective states in social interactions can be used through social media to analyze and support lifestyle behaviour. A computational model is provided that integrates both mechanisms for the impact of one's emotions on behaviour, and for the impact of emotions of others on one's own emotion. The model is used to assess the state of a user with regard to a lifestyle goal (such as exercising frequently), based on extracted information of emotions exchanged in social interaction. Support is provided by proposing ways to affect these social interactions, which will indirectly influence the impact of the emotions of others. An ambient intelligent system based on this model has been implemented for the social medium Twitter.
Keywords: Social media; emotion; lifestyle support; Ambient Intelligence
A Computational Analysis of Joint Decision Making Processes BIBAFull-Text 292-308
  Rob Duell; Jan Treur
In this paper a computational analysis is made of the circumstances under which joint decisions are or are not reached. Joint decision making as considered does not only concern a choice for a common decision option, but also a good feeling about it, and mutually acknowledged empathic understanding. As a basis a computational social agent model for joint decision making is used. The model was inspired by principles from neurological theories on mirror neurons, internal simulation, and emotion-related valuing. The computational analysis determines the different possible outcomes of joint decision making processes, and the types of processes leading to these outcomes.
Collaboratively Constructing a VDL-Based Icon System for Knowledge Tagging BIBAKFull-Text 309-322
  Xiaoyue Ma; Jean-Pierre Cahier
Tag system for a knowledge organization system centralizes and provides the tags that can be employed in classifying, sharing and seeking knowledge for personal or organizational use within a social community. Considering current constraints of textual tag system and developing iconic tag system, VDL-based iconic tag system has been built and validated to improve knowledge tagging with symbolic interpretation and graphical organization of tag structure. In this paper, we are proposing cooperative creation of such special icon system where VDL-based icons will be applied for social knowledge tagging and sharing. This VDL-based icon system could also serve as a visual knowledge organization system to facilitate icon searching in a given context.
Keywords: icon; tag system; knowledge tagging; knowledge organization system; Hypertopic; social community
A Multi-dimensional and Event-Based Model for Trust Computation in the Social Web BIBAKFull-Text 323-336
  Barbara Carminati; Elena Ferrari; Marco Viviani
In this paper, we propose a general-purpose Trust Layer that fits and exploits the emerging concept of Social Web. Key features of our proposal are the consideration of several dimensions for trust computation and the exploitation of social interaction dynamics over the Web, through the definition and the evaluation of event patterns and trust rules. Besides presenting our trust model, we discuss a case study on the ACM Digital Library social environment.
Keywords: Trust; Multi-dimensional Social Trust; Social Web
On Recommending Hashtags in Twitter Networks BIBAKFull-Text 337-350
  Su Mon Kywe; Tuan-Anh Hoang; Ee-Peng Lim; Feida Zhu
Twitter network is currently overwhelmed by massive amount of tweets generated by its users. To effectively organize and search tweets, users have to depend on appropriate hashtags inserted into tweets. We begin our research on hashtags by first analyzing a Twitter dataset generated by more than 150,000 Singapore users over a three-month period. Among several interesting findings about hashtag usage by this user community, we have found a consistent and significant use of new hashtags on a daily basis. This suggests that most hashtags have very short life span. We further propose a novel hashtag recommendation method based on collaborative filtering and the method recommends hashtags found in the previous month's data. Our method considers both user preferences and tweet content in selecting hashtags to be recommended. Our experiments show that our method yields better performance than recommendation based only on tweet content, even by considering the hashtags adopted by a small number (1 to 3) of users who share similar user preferences.
Keywords: Twitter; hashtag; recommendation systems
A Framework for the Design and Synthesis of Coordinated Social Systems BIBAFull-Text 351-364
  Wynn Stirling; Christophe Giraud-Carrier; Teppo Felin
This paper describes how a nascent collective of individuals can coalesce into a complex social system. The systematic study of such scenarios requires a mathematical framework within which to model the behavior of the individual members of the collective. As individuals interact, they develop social relationships and exchange resources -- that is, they develop social capital that quantifies the value of social influence that individuals exert on each other. Social capital can be expressed via conditional preference orderings for each individual. Conditional preferences reflect the influence relationships of an interacting social collective. Conditional preference orderings can then be aggregated via conditional game theory to form a concordant utility that provides an emergent group-level ordering of the harmony of interests of the members of the collective. We can thus develop a complete social model that takes into consideration all social relationships as they propagate through the system. Solution concepts can then be defined that simultaneously account for both group-level and individual-level interests.
Swayed by Friends or by the Crowd? BIBAFull-Text 365-378
  Zeinab Abbassi; Christina Aperjis; Bernardo A. Huberman
We have conducted three empirical studies of the effects of friend recommendations and general ratings on how online users make choices. We model and quantify how a user deciding between two choices trades off an additional rating star with an additional friend's recommendation when selecting an item. We find that negative opinions from friends are more influential than positive opinions, and people exhibit "more random" behavior in their choices when the decision involves less cost and risk. Our results are quite general in the sense that people across different demographics trade off recommendations from friends and ratings from the general public in a similar fashion.
Are Twitter Users Equal in Predicting Elections? A Study of User Groups in Predicting 2012 U.S. Republican Presidential Primaries BIBAKFull-Text 379-392
  Lu Chen; Wenbo Wang; Amit P. Sheth
Existing studies on predicting election results are under the assumption that all the users should be treated equally. However, recent work [14] shows that social media users from different groups (e.g., "silent majority" vs. "vocal minority") have significant differences in the generated content and tweeting behavior. The effect of these differences on predicting election results has not been exploited yet. In this paper, we study the spectrum of Twitter users who participate in the on-line discussion of 2012 U.S. Republican Presidential Primaries, and examine the predictive power of different user groups (e.g., highly engaged users vs. lowly engaged users, right-leaning users vs. left-leaning users) against Super Tuesday primaries in 10 states. The insights gained in this study can shed light on improving the social media based prediction from the user sampling perspective and more.
Keywords: Electoral Prediction; Twitter Analytics; Social Intelligence; User Categorization; Engagement Degree; Tweet Mode; Content Type; Political Preference
Web Page Recommendation Based on Semantic Web Usage Mining BIBAKFull-Text 393-405
  Soheila Abrishami; Mahmoud Naghibzadeh; Mehrdad Jalali
The growth of the web has created a big challenge for directing the user to the Web pages in their areas of interest. Meanwhile, web usage mining plays an important role in finding these areas of interest based on user's previous actions. The extracted patterns in web usage mining are useful in various applications such as recommendation. Classical web usage mining does not take semantic knowledge and content into pattern generations. Recent researches show that ontology, as background knowledge, can improve pattern's quality. This work aims to design a hybrid recommendation system based on integrating semantic information with Web usage mining and page clustering based on semantic similarity. Since the Web pages are seen as ontology individuals, frequent navigational patterns are in the form of ontology instances instead of Web page addresses, and page clustering is done using semantic similarity. The result is used for generating web page recommendations to users. The recommender engine presented in this paper which is based on semantic patterns and page clustering, creates a list of appropriate recommendations. The results of the implementation of this hybrid recommendation system indicate that integrating semantic information and page access sequence into the patterns yields more accurate recommendations.
Keywords: Web usage mining; semantic web; ontology; web page recommendation; page clustering
Scalable Analysis for Large Social Networks: The Data-Aware Mean-Field Approach BIBAFull-Text 406-419
  Julie M. Birkholz; Rena Bakhshi; Ravindra Harige; Maarten van Steen; Peter Groenewegen
Studies on social networks have proved that endogenous and exogenous factors influence dynamics. Two streams of modeling exist on explaining the dynamics of social networks: 1) models predicting links through network properties, and 2) models considering the effects of social attributes. In this interdisciplinary study we work to overcome a number of computational limitations within these current models. We employ a mean-field model which allows for the construction of a population-specific model informed from empirical research for predicting links from both network and social properties in large social networks.. The model is tested on a population of conference coauthorship behavior, considering a number of parameters from available Web data. We address how large social networks can be modeled preserving both network and social parameters. We prove that the mean-field model, using a data-aware approach, allows us to overcome computational burdens and thus scalability issues in modeling large social networks in terms of both network and social parameters. Additionally, we confirm that large social networks evolve through both network and social-selection decisions; asserting that the dynamics of networks cannot singly be studied from a single perspective but must consider effects of social parameters.
A Survey of Recommender Systems in Twitter BIBAKFull-Text 420-433
  Su Mon Kywe; Ee-Peng Lim; Feida Zhu
Twitter is a social information network where short messages or tweets are shared among a large number of users through a very simple messaging mechanism. With a population of more than 100M users generating more than 300M tweets each day, Twitter users can be easily overwhelmed by the massive amount of information available and the huge number of people they can interact with. To overcome the above information overload problem, recommender systems can be introduced to help users make the appropriate selection. Researchers have began to study recommendation problems in Twitter but their works usually address individual recommendation tasks. There is so far no comprehensive survey for the realm of recommendation in Twitter to categorize the existing works as well as to identify areas that need to be further studied. The paper therefore aims to fill this gap by introducing a taxonomy of recommendation tasks in Twitter, and to use the taxonomy to describe the relevant works in recent years. The paper further presents the datasets and techniques used in these works. Finally, it proposes a few research directions for recommendation tasks in Twitter.
Keywords: Twitter; recommender systems; personalization
A Multi-view Content-Based User Recommendation Scheme for Following Users in Twitter BIBAKFull-Text 434-447
  Milen Chechev; Petko Georgiev
This paper describes recommendation techniques that help users to find potentially interesting people to follow at Twitter. The explored techniques are based on a confirmed assumption that the recent activity of users is indicative of their latest friend preferences. Several content-based recommendation strategies are explored, compared and tested. Among them the foundations for a novel hybridization framework are provided and a multi-view approach towards modeling user profiles is considered. The training and test database is crawled with real users and tweets from the Twitter network. A non-standard evaluation scheme is applied in an offline testing context for the various algorithms. Conclusions are drawn as to the viability, relative predictive power and accuracy of the recommendation approaches.
Keywords: recommendation system; social network; Twitter; content-based; multi-view hybridization
Spam Fighting in Social Tagging Systems BIBAKFull-Text 448-461
  Sasan Yazdani; Ivan Ivanov; Morteza AnaLoui; Reza Berangi; Touradj Ebrahimi
Tagging in online social networks is very popular these days, as it facilitates search and retrieval of diverse resources available online. However, noisy and spam annotations often make it difficult to perform an efficient search. Users may make mistakes in tagging and irrelevant tags and resources may be maliciously added for advertisement or self-promotion. Since filtering spam annotations and spammers is time-consuming if it is done manually, machine learning approaches can be employed to facilitate this process. In this paper, we propose and analyze a set of distinct features based on user behavior in tagging and tags popularity to distinguish between legitimate users and spammers. The effectiveness of the proposed features is demonstrated through a set of experiments on a dataset of social bookmarks.
Keywords: Social tagging systems; Social spam; Spam detection; Spammers; User behavior; Tags popularity
The Multidimensional Study of Viral Campaigns as Branching Processes BIBAKFull-Text 462-474
  Jaroslaw Jankowski; Radoslaw Michalski; Przemyslaw Kazienko
Viral campaigns on the Internet may follow variety of models, depending on the content, incentives, personal attitudes of sender and recipient to the content and other factors. Due to the fact that the knowledge of the campaign specifics is essential for the campaign managers, researchers are constantly evaluating models and real-world data. The goal of this article is to present the new knowledge obtained from studying two viral campaigns that took place in a virtual world which followed the branching process. The results show that it is possible to reduce the time needed to estimate the model parameters of the campaign and, moreover, some important aspects of time-generations relationship are presented.
Keywords: viral campaigns; diffusion of information; branching process; social network analysis; virtual worlds
Models of Social Groups in Blogosphere Based on Information about Comment Addressees and Sentiments BIBAKFull-Text 475-488
  Bogdan Gliwa; Jaroslaw Kozlak; Anna Zygmunt; Krzysztof Cetnarowicz
This work concerns the analysis of number, sizes and other characteristics of groups identified in the blogosphere using a set of models identifying social relations. These models differ regarding identification of social relations, influenced by methods of classifying the addressee of the comments (they are either the post author or the author of a comment on which this comment is directly addressing) and by a sentiment calculated for comments considering the statistics of words present and connotation. The state of a selected blog portal was analyzed in sequential, partly overlapping time intervals. Groups in each interval were identified using a version of the CPM algorithm, on the basis of them, stable groups, existing for at least a minimal assumed duration of time, were identified.
Keywords: social network analysis; groups; blogosphere; sentiment
Note: Best paper award
Dark Retweets: Investigating Non-conventional Retweeting Patterns BIBAKFull-Text 489-502
  Norhidayah Azman; David E. Millard; Mark J. Weal
Retweets are an important mechanism for recognising propagation of information on the Twitter social media platform. However, many retweets do not use the official retweet mechanism, or even community established conventions, and these "dark retweets" are not accounted for in many existing analysis. In this paper, a comprehensive matrix of tweet propagation is presented to show the different nuances of retweeting, based on seven characteristics: whether it is proprietary, the mechanism used, whether it is directed to followers or non-followers, whether it mentions other users, if it is explicitly propagating another tweet, if it links to an original tweet, and what is the audience it is pushed to. Based on this matrix and two assumptions of retweetability, the degrees of a retweet's "darkness" can be determined. This matrix was evaluated over 2.3 million tweets and it was found that dark retweets amounted to 12.86% (for search results less than 1500 tweets per URL) and 24.7% (for search results including more than 1500 tweets per URL) respectively. By extrapolating these results with those found in existing studies, potentially thousands of retweets may be hidden from existing studies on retweets.
Keywords: retweets; tweet propagation; dark retweets; Twitter; microblogs
Studying Paths of Participation in Viral Diffusion Process BIBAKFull-Text 503-516
  Jaroslaw Jankowski; Sylwia Ciuberek; Anita Zbieg; Radoslaw Michalski
Authors propose a conceptual model of participation in viral diffusion process composed of four stages: awareness, infection, engagement and action. To verify the model it has been applied and studied in the virtual social chat environment settings. The study investigates the behavioural paths of actions that reflect the stages of participation in the diffusion and presents shortcuts, that lead to the final action -- the attendance in a virtual event. The results show that the participation in each stage of the process increases the probability of reaching the final action. Nevertheless, the majority of users involved in the virtual event did not go through each stage of the process but followed the shortcuts. That suggests that the viral diffusion process is not necessarily a linear sequence of human actions but rather a dynamic system.
Keywords: information diffusion; online social networks; participation model; multistage analysis
Paradox of Proximity -- Trust and Provenance within the Context of Social Networks and Policy BIBAKFull-Text 517-530
  Somya Joshi; Timo Wandhöfer; Vasilis Koulolias; Catherine Van Eeckhaute; Beccy Allen; Steve Taylor
With social networks evolving and integrating within traditional policy domains, the question arises -- do we have in our hands a tool for genuine participation, transparency and dialogue, or are the concerns surrounding privacy, trust, provenance and localization still haunting and shaping the arena? In this paper, we discuss this very question via the illustrative lens of the WeGov Project. We start by providing a critical rethinking of e-governance within the context of social media. We then move onto an in depth look at the WeGov project, its toolkit, end-user engagement strategies and methodologies. Finally we draw from our findings some critical insights into the impacts on and implications of such technologies for the policy-making environment. We conclude with a set of recommendations for future work in this area as well as a summary of key lessons learnt within this innovative initiative.
Keywords: Trust; Provenance; Social Networks
Namelings BIBAKFull-Text 531-534
  Folke Mitzlaff; Gerd Stumme
During the exhausting search for a given name for the yet unborn baby, the idea of a name recommendation system based on relations mined from the "social web" was born. This demonstration paper presents the Nameling, a recommendation system, search engine and academic research platform for given names, which attracted more than 30,000 users within four months, underpinning the relevance of the task and associated research questions.
Keywords: Given Names; Network Analysis; Recommendation System
SocialTrends: A Web Application for Monitoring and Visualizing Users in Social Media BIBAKFull-Text 535-538
  Maurizio Tesconi; Davide Gazzé; Angelica Lo Duca
Nowadays social media trends are becoming very important to describe the variation of popularity, activity and influence of an entity. In this paper we define an abstract model which can be used on different social media to compare metrics with the same meaning. In particular we describe three classes of metrics: popularity, activity and influence. We also present SocialTrends, a web application (http://www.social-trends.it/) which collects, elaborates and visualizes social media data. Finally, we describe one experiment we have done to test SocialTrends.
Keywords: Social Media Visualization; Monitoring; Trends; Data Gathering
Demonstration of Dynamic Targeting in an Online Social Medium BIBAKFull-Text 539-542
  Peter Laflin; Fiona Ainley; Amanda Otley; Alexander V. Mantzaris; Desmond J. Higham
A novel way of calculating online influence has been proposed in [2,1]. Bloom Agency have created new online software capable of collecting social data and calculating these new influence metrics in real time. A demonstration of this software will be given at the conference. Delegates will be encouraged to Tweet using the #socinfo2012 hashtag and the influence of the top ten Tweeters will be shown, along with a visualisation of the evolving conversation.
Keywords: Twitter; Social Media; Online Influence; Dynamic Networks; Katz Centrality
Navigating between Chaos and Bureaucracy: Backgrounding Trust in Open-Content Communities BIBAKFull-Text 543-557
  Paul B. de Laat
Many virtual communities that rely on user-generated content (such as social news sites, citizen journals, and encyclopedias in particular) offer unrestricted and immediate 'write access' to every contributor. It is argued that these communities do not just assume that the trust granted by that policy is well-placed; they have developed extensive mechanisms that underpin the trust involved ('backgrounding'). These target contributors (stipulating legal terms of use and developing etiquette, both underscored by sanctions) as well as the contents contributed by them (patrolling for illegal and/or vandalist content, variously performed by humans and bots; voting schemes). Backgrounding trust is argued to be important since it facilitates the avoidance of bureaucratic measures that may easily cause unrest among community members and chase them away.
Keywords: access; bots; bureaucracy; etiquette; open content; trust; vandalism; Wikipedia