HCI Bibliography Home | HCI Conferences | UMAP Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
UMAP Tables of Contents: 00010203040506070809101112131415

Proceedings of AH 2004 Adaptive Hypermedia and Adaptive Web-based Systems 2004-08-23

Fullname:Proceedings of the 3rd International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems
Editors:Paul M. E. De Bra; Wolfgang Nejdl
Location:Eindhoven, Netherlands
Dates:2004-Aug-23 to 2004-Aug-26
Publisher:Springer-Verlag
Series:Lecture Notes in Computer Science, 2004, Volume 3137
Standard No:ISBN: 978-3-540-22895-0 (Print) 978-3-540-27780-4 (Online); hcibib: UMAP04
Papers:69
Pages:440
Links:Online Proceedings
  1. Keynote Speakers (Abstracts)
  2. Full Papers
  3. Short Papers
  4. Doctoral Consortium
  5. Posters

Keynote Speakers (Abstracts)

Ambient Intelligence BIBAFull-Text 1
  Emile Aarts
In the near future our homes will have a distributed network of intelligent devices that provides us with information, communication, and entertainment. Furthermore, these systems will adapt themselves to the user and even anticipate on user needs. These consumer systems will differ substantially from contemporary equipment through their appearance in peoples' environments, and through the way users interact with them. Ambient Intelligence is the term that is used to denote this new paradigm for in-home computing and entertainment. Salient features of this new concept are ubiquitous computing, and natural interaction. Recent developments in technology, the Internet, the consumer electronics market, and social developments indicate that this dream might become reality soon. First prototypes of ambient intelligent home systems have been developed, but the realization of true ambient intelligence calls for much additional research of multidisciplinary teams consisting of technologists, designers, and human behavior scientists.
   The presentation will elaborate on some of these aspects and show where we are in the development of ambient intelligence.
Collaborative Agents for 2D Interfaces and 3D Robots BIBAFull-Text 2
  Candace L. Sidner
In this talk I will discuss our several years experience in systems for collaborative interface agents. I will discuss the tool called COLLAGEN(tm) for COLLaborative AGENts. COLLAGEN(tm) is a Java middleware system, that makes it possible to build an agent with a rich model of conversation and collaboration for a set of tasks with a user for an application, provided by the developer using Collagen. I will demo two of the many systems built in our lab and elsewhere, which rely on COLLAGEN(tm): one with speech for a desktop application, and one for a 3D robot (via videclips). I will discuss the way Collagen was developed from theories of conversation and collaboration, how plan recognition plays a role in COLLAGEN(tm), and I will point out new directions of our work, especially given the nonverbal gestural abilities of our humanoid robot.
A Curse of Riches or a Blessing? Information Access and Awareness Under Scarce Cognitive Resources BIBAFull-Text 3
  Eric Horvitz
The informational landscape of the world has been changing quickly. The fast-paced drop in the cost of storage and bandwidth over the last decade, coupled with the rapid expansion in the number of content sources, has made unprecedented quantities of information available to people. Beyond external sources of content, typical personal stores now rival the size of the entire Web just a short time ago. But we face a challenging bottleneck: In stark contrast to the explosive growth in public and private content, stands our limited time-and unchanging abilities. For increasing numbers of people in the world, the primary informational bottleneck is our scarce attentional and memory resources. I will present research on addressing such informational bottlenecks with tools for searching, browsing, remembering, and staying aware. I will review challenges and opportunities of employing automated learning and reasoning methods, including efforts to construct and leverage models of attention and memory. Finally, I will touch on the promise of developing new designs for interaction and display informed by psychological findings on visual attention and spatial memory.

Full Papers

Supporting Metadata Creation with an Ontology Built from an Extensible Dictionary BIBAFull-Text 4-13
  Trent Apted; Judy Kay; Andrew Lum
This paper describes Metasaur, which supports creation of metadata about the content of learning objects. The core of Metasaur is a visualisation for an ontology of the domain. We describe how we build lightweight ontologies for Metasaur automatically from existing dictionaries and how a user can enhance the ontology with additional terms. We report our use of Metasaur to mark up a set of audio lecture learning objects for use in a course.
Interaction with Web Services in the Adaptive Web BIBAFull-Text 14-23
  Liliana Ardissono; Anna Goy; Giovanna Petrone; Marino Segnan
The provision of personalized services based on the orchestration of simpler Web Services is often viewed as an activity that can be performed in an automated way, without involving the end-user. This paper addresses the need to involve the user in the loop and discusses the communication challenges imposed by this viewpoint. The paper also presents a conversation model for the management of the communication between Web Service consumers and providers aimed at addressing those challenges.
Social Adaptive Navigation Support for Open Corpus Electronic Textbooks BIBAFull-Text 24-33
  Peter Brusilovsky; Girish Chavan; Rosta Farzan
Closed corpus AH systems demonstrate what is possible to achieve with adaptive hypermedia technologies; however they are impractical for dealing with the large volume of open corpus resources. Our system, Knowledge Sea II, presented in this paper explores social adaptive navigation support, an approach for providing personalized guidance in the open corpus context. Following the ideas of social navigation, we have attempted to organize a personalized navigation support that is based on past learners' interaction with the system. The social adaptive navigation support implemented in our system was considered quite useful by students participating in the classroom study of Knowledge Sea II. At the same time, some user comments indicated the need to provide more powerful navigation support.
PROS: A Personalized Ranking Platform for Web Search BIBAFull-Text 34-43
  Paul-Alexandru Chirita; Daniel Olmedilla; Wolfgang Nejdl
Current search engines rely on centralized page ranking algorithms which compute page rank values as single (global) values for each Web page. Recent work on topic-sensitive PageRank [6] and personalized PageRank [8] has explored how to extend PageRank values with personalization aspects. To achieve personalization, these algorithms need specific input: [8] for example needs a set of personalized hub pages with high PageRank to drive the computation. In this paper we show how to automate this hub selection process and build upon the latter algorithm to implement a platform for personalized ranking. We start from the set of bookmarks collected by a user and extend it to contain a set of hubs with high PageRank related to them. To get additional input about the user, we implemented a proxy server which tracks and analyzes user's surfing behavior and outputs a set of pages preferred by the user. This set is then enrichened using our HubFinder algorithm, which finds related pages, and used as extended input for the [8] algorithm. All algorithms are integrated into a prototype of a personalized Web search system, for which we present a first evaluation.
A P2P Distributed Adaptive Directory BIBAFull-Text 44-54
  Gennaro Cordasco; Vittorio Scarano; Cristiano Vitolo
We describe a P2P system that offers a distributed, cooperative and adaptive environment for bookmark sharing. DAD offers an adaptive environment since it provides suggestions about the navigation based on (a) the bookmarks, (b) the feedback implicitly provided by users and (c) the structure of the Web. Our system is fully scalable because of its peer-to-peer architecture and provides, also, an infrastructure to build easily P2P overlay networks.
Developing Active Learning Experiences for Adaptive Personalised eLearning BIBAFull-Text 55-64
  Declan Dagger; Vincent P. Wade; Owen Conlan
Developing adaptive, rich-media, eLearning courses tends to be a complex, highly-expensive and time-consuming task. A typical adaptive eLearning course will involve a multi-skilled development team of technologists, instructional developers, subject matter experts and integrators. Even where the adaptive course attempts to reuse existing digital resources, considerable effort is still required in the integration of the adaptive techniques and curriculum. This paper tackles the fundamental challenges of extending adaptivity across not only content (based on prior knowledge, goals, learning styles, connectivity etc.) but also across adaptive pedagogic approaches, communication tools and a range of e-activity types which are required for effective, deeper learning. This paper identifies key activities and requirements for adaptive course construction and presents the design of a tool to allow the rapid construction of such courses. The paper outlines the usage of this tool in the form of a case study and presents its research findings.
Adaptive User Modeling for Personalization of Web Contents BIBAFull-Text 65-74
  Alberto Díaz; Pablo Gervás
This paper presents a system for personalization of web contents based on a user model that stores long term and short term interests. Long term interests are modeled through the selection of specific and general categories, and keywords for which the user needs information. However, user needs change over time as a result of his interaction with received information. For this reason, the user model must be capable of adapting to those shifts in interest. In our case, this adaptation of the user model is performed by a short term model obtained from user provided feedback. The evaluation performed with 100 users during 15 days has determined that the combined use of long and short term models performs best when specific and general categories and keywords are used together for the long term model.
Invoking Web Applications from Portals: Customisation Implications BIBAFull-Text 75-84
  Oscar Díaz; Iñaki Paz
Customisation sits at the core of current portal technology. So does content syndication as well as the most recent, application syndication whereby external applications can be integrated into the portal realm through the use of Portlet technology. However, these applications do not always exhibit the sophisticated customisation mechanisms available within a portal. This leads to a discontinuity in the user experience when accessing an external application being syndicated within a portal. This work introduces the notion of bridge Portlet as a proxy to an external Web application. This Portlet is responsible for customising the external application to the portal environment, supplementing it with commodities such as single sign on, profiles and the like. The paper illustrates how to improve the adaptation of the external application through a bookmarking module enhancement. Now, portal users can enjoy bookmarking facilities not only for portal-oriented applications but also for external Web applications.
The Personal Reader: Personalizing and Enriching Learning Resources Using Semantic Web Technologies BIBAKFull-Text 85-94
  Peter Dolog; Nicola Henze; Wolfgang Nejdl; Michael Sintek
Traditional adaptive hypermedia systems have focused on providing adaptation functionality on a closed corpus, while Web search interfaces have delivered non-personalized information to users. In this paper, we show how we integrate closed corpus adaptation and global context provision in a Personal Reader environment. The local context consists of individually optimized recommendations to learning materials within the given corpus; the global context provides individually optimized recommendations to resources found on the Web, e. g., FAQs, student exercises, simulations, etc. The adaptive local context of a learning resource is generated by applying methods from adaptive educational hypermedia in a semantic web setting. The adaptive global context is generated by constructing appropriate queries, enrich them based on available user profile information, and, if necessary, relax them during the querying process according to available metadata.
Keywords: adaptive hypermedia; personalization; adaptive web; semantic web; reasoning rules; querying the semantic web
An Experiment in Social Search BIBAFull-Text 95-103
  Jill Freyne; Barry Smyth
Social search is an approach to Web search that attempts to offer communities of like-minded individuals more targeted search services, based on the search behaviour of their peers, bringing together ideas from Web search, social networking and personalization. In this paper we describe the I-SPY architecture for social search and present the results of a recent live-user evaluation that highlight the potential benefits of our approach in a realistic search setting.
Recent Soft Computing Approaches to User Modeling in Adaptive Hypermedia BIBAFull-Text 104-114
  Enrique Frías-Martínez; George Magoulas; Sherry Chen; Robert Macredie
The ability of an adaptive hypermedia system to create tailored environments depends mainly on the amount and accuracy of information stored in each user model. One of the difficulties that user modeling faces is the necessity of capturing the imprecise nature of human behavior. Soft Computing has the ability to handle and process uncertainty which makes it possible to model and simulate human decision-making. This paper surveys different soft computing techniques that can be used to efficiently and accurately capture user behavior. The paper also presents guidelines that show which techniques should be used according to the task implemented by the application.
A Flexible Composition Engine for Adaptive Web Sites BIBAKFull-Text 115-125
  Serge Garlatti; Sébastien Iksal
Nowadays, adaptive web sites must have the ability to use distributed repositories. The variety of available data sources and user profiles quickly lead to a combinatorial explosion of web site versions. It is impossible to manage these web sites without some form of automation. We are interested in web sites for which users belong to a kind of community of practices: they share a common knowledge to work together. We claim that the explicit knowledge of communities is the key issue to automate the hypermedia generation and thus to ensure consistency and comprehension. We have designed a flexible adaptive composition engine. In our framework, hypermedia consistency is managed by an author through content, adaptation and sites structure at knowledge level and based on his know-how. Our major contribution consists of: (i) a semantic organization of resources, (ii) a declarative specification of adaptation and (iii) the flexibility of the composition engine.
Keywords: Adaptation; Virtual Document; Composition Engine; Semantic Web; Metadata; User Model
Intelligent Support to the Retrieval of Information About Hydric Resources BIBAFull-Text 126-135
  Cristina Gena; Liliana Ardissono
This paper presents the adaptive search features offered by ACQUA, a Web-based system presenting information about the hydric resources of the Piedmont Italian Region. ACQUA enables the user to retrieve qualitative and quantitative data in different formats, supporting direct data manipulation. Moreover, the system supports the user in the search for information by complementing her explicit search queries with follow-up queries frequently occurring together in navigation paths. In this way, the user may retrieve complete information in an efficient way. The paper describes the results of an evaluation of the adaptivity features carried out with real users.
CUMAPH: Cognitive User Modeling for Adaptive Presentation of Hyper-documents. An Experimental Study BIBAFull-Text 136-145
  Halima Habieb-Mammar; Franck Tarpin-Bernard
In this paper we present the CUMAPH environment (Cognitive User Modeling for Adaptive Presentation of Hyper-document). The aim of this environment is to adapt hyper-document presentation to the cognitive user profile. The architecture of this environment is based on four main components: a cognitive user model, a hyper-document generation process, an adaptive process and a generic style sheet. These components and their combinations are described in detail. To validate our approach, an experimental study was conducted upon a population of 50 students. The experimental steps and the main results obtained are discussed in this paper.
Personalized Web Advertising Method BIBAFull-Text 146-155
  Przemyslaw Kazienko; Michal Adamski
Personalization of online advertising is a great challenge while the market is moving and adapting to the realities of the Internet. Many existing approaches to advertisement recommendation are based on demographic targeting or on information gained directly from the user. In this paper we introduce the AD ROSA system for automatic web banner personalization, which integrates web usage and content mining techniques to reduce user input and to respect the user's privacy. Furthermore, the advertising campaign policy, an important factor for both the publisher and advertiser, is taken into consideration. To enable online personalized advertising the integration of all relevant information is performed in one vector space.
Flexible Navigation Support in the WINDS Learning Environment for Architecture and Design BIBAFull-Text 156-165
  Milos Kravcik; Marcus Specht
The paper presents the knowledge structure of the WINDS system and shows the implementation of its learning environment, which is adaptive and adaptable. It supports different learning approaches and gives the learner guidance by coaching. The content of the WINDS virtual university is structured in SCORM compliant learning objects and connected with a semantic layer of learning concepts. The usage of this structure in the ALE learning environment is described and results from a first evaluation study are reported.
Evaluation of WINDS Authoring Environment BIBAFull-Text 166-175
  Milos Kravcik; Marcus Specht; Reinhard Oppermann
Authoring tools for adaptive educational hypermedia are still rarely available for a wider public. In the WINDS project, we have developed the Adaptive Learning Environment (ALE) for various European universities active in the area of design and architecture. Teachers without programming skills have created 21 courses in the ALE authoring environment, which simplifies the process providing learning object templates and enabling reusability of materials. This paper describes the WINDS authoring approach and presents some evaluation results.
On the Dynamic Generation of Compound Critiques in Conversational Recommender Systems BIBAFull-Text 176-184
  Kevin McCarthy; James Reilly; Lorraine McGinty; Barry Smyth
Conversational recommender systems help to guide users through a product-space towards a particular product that meets their specific requirements. During the course of a "conversation" with the user the recommender system will suggest certain products and use feedback from the user to refine future suggestions. Critiquing has proven to be a powerful and popular form of feedback. Critiques allow the user to express a preference over part of the feature-space; for example, in a vacation/travel recommender a user might indicate that they are looking for a "less expensive" vacation than the one suggested, thereby critiquing the price feature. Usually the set of critiques that the user can chose from is fixed as part of the basic recommender interface. In this paper we will propose a more dynamic critiquing approach where high-quality critiques are automatically generated during each recommendation cycle from the remaining product-cases. We show that these dynamic critiques can lead to more efficient recommendation performance by helping the user to more rapidly focus in on the right region of the product-space.
Evaluating Adaptive Problem Selection BIBAFull-Text 185-194
  Antonija Mitrovic; Brent Martin
This paper presents an evaluation study that compares two different problem selection strategies for an Intelligent Tutoring System (ITS). The first strategy uses static problem complexities specified by the teacher to select problems that are appropriate for a student based on his/her current level of ability. The other strategy is more adaptive: individual problem difficulties are calculated for each student based on the student's specific knowledge, and the appropriate problem is then selected based on these dynamic difficulty measures. The study was performed in the context of the SQL-Tutor system. The results show that adaptive problem selection based on dynamically generated problem difficulties can have a positive effect on student learning performance.
Adaptive Presentation and Navigation for Geospatial Imagery Tasks BIBAFull-Text 195-204
  Dympna O'Sullivan; Eoin McLoughlin; Michela Bertolotto; David C. Wilson
Advances in technology for digital image capture have given rise to information overload problems in the geosciences and fields that rely on geospatial image retrieval. To help address such imagery information overload, our research is developing methods to extract and apply contextual knowledge relating user task goals to the images being used. As users analyze imagery retrieved to support specific tasks, multiple relevant images and salient image content can be captured together with task annotations to provide a basis for contextual knowledge management support. This paper describes how our environment for image interaction leverages captured task-based knowledge to support adaptive presentation and navigation in the space of available imagery.
Myriad: An Architecture for Contextualized Information Retrieval and Delivery BIBAFull-Text 205-214
  Cécile Paris; Mingfang Wu; Keith Vander Linden; Matthew Post; Shijian Lu
Users' information needs are largely driven by the context in which they make their decisions. This context is dynamic. It includes the users' characteristics, their current domain of application, the tasks they commonly perform and the device they are currently using. This context is also evolving. When one information need is satisfied, another is likely to emerge. An information access system must, therefore, be able to track this dynamic and evolving context, and exploit it to retrieve actionable information from appropriate sources and deliver it in a form suitable for the current situation. This paper presents a generic architecture that supports the construction of information retrieval and delivery systems that make use of context. The architecture, called Myriad, includes an adaptive virtual document planner, and explicit, dynamic representations of the user's current context.
Cross-Media and Elastic Time Adaptive Presentations: The Integration of a Talking Head Tool into a Hypermedia Formatter BIBAFull-Text 215-224
  Rogério Ferreira Rodrigues; Paula Salgado Lucena Rodrigues; Bruno Feijó; Luiz Velho
This paper describes the integration of a facial animation tool (Expressive Talking Heads -- ETHs) with an adaptive hypermedia formatter (HyperProp formatter). This formatter is able to adjust document presentations based on the document temporal constraints (e.g. synchronization relationships), the presentation platform parameters (e.g. available bandwidth and devices), and the user profile (e.g. language, accessibility, etc.). This work describes how ETHs augments the capability for creating adaptive hypermedia documents with HyperProp formatter. The paper also presents the adaptation facilities offered by the main hypermedia language (Nested Context Language -- NCL) HyperProp system works with, and details the implementation extensions of Expressive Talking Heads that turned it an adaptive presentation tool.
Assessing Cognitive Load in Adaptive Hypermedia Systems: Physiological and Behavioral Methods BIBAFull-Text 225-234
  Holger Schultheis; Anthony Jameson
It could be advantageous in many situations for an adaptive hypermedia system to have information about the cognitive load that the user is currently experiencing. A literature review of the methods proposed to assess cognitive load reveals: (1) that pupil size seems to be one of the most promising indicators of cognitive load in applied contexts and (2) that its suitability for use as an on-line index in everyday situations has not yet been tested adequately. Therefore, the aim of the present study was to evaluate the usefulness of the pupil size index in such situations. To this end, pupil diameter and event-related brain potentials were measured while subjects read texts of different levels of difficulty. As had been hypothesized, more difficult texts led to lower reading speed, higher subjective load ratings, and a reduced P300 amplitude. But text difficulty, surprisingly, had no effect on pupil size. These results indicate that pupil size may not be suitable as an index of cognitive load for adaptive hypermedia systems. Instead, behavioral indicators such as reading speed may be more suitable.
Context-Aware Recommendations in the Mobile Tourist Application COMPASS BIBAFull-Text 235-244
  Mark van Setten; Stanislav Pokraev; Johan Koolwaaij
This paper describes the context-aware mobile tourist application COMPASS that adapts its services to the user's needs based on both the user's interests and his current context. In order to provide context-aware recommendations, a recommender system has been integrated with a context-aware application platform. We describe how this integration has been accomplished and how users feel about such an adaptive tourist application.
Utilizing Artificial Learners to Help Overcome the Cold-Start Problem in a Pedagogically-Oriented Paper Recommendation System BIBAFull-Text 245-254
  Tiffany Tang; Gordon McCalla
In this paper we discuss the cold-start problem in an evolvable paper recommendation e-learning system. We carried out an experiment using artificial and human learners at the same time. Artificial learners are used to solve the cold-start recommendation problem when no paper has been rated by the learners. Experimental results are encouraging, showing that using artificial learners achieves better performance in terms of learner subjective ratings; and more importantly, human learners are satisfied with the recommendations received.
Unison-CF: A Multiple-Component, Adaptive Collaborative Filtering System BIBAKFull-Text 255-264
  Manolis Vozalis; Konstantinos G. Margaritis
In this paper we present the Unison-CF algorithm, which provides an efficient way to combine multiple collaborative filtering approaches, drawing advantages from each one of them. Each collaborative filtering approach is treated as a separate component, allowing the Unison-CF algorithm to be easily extended. We evaluate the Unison-CF algorithm by applying it on three existing filtering approaches: User-based Filtering, Item-based Filtering and Hybrid-CF. Adaptation is utilized and evaluated as part of the filtering approaches combination. Our experiments show that the Unison-CF algorithm generates promising results in improving the accuracy and coverage of the existing filtering algorithms.
Keywords: collaborative filtering; memory-based filtering; adaptation; personalization; prediction; recommender systems
Using SiteRank for Decentralized Computation of Web Document Ranking BIBAKFull-Text 265-274
  Jie Wu; Karl Aberer
The PageRank algorithm demonstrates the significance of the computation of document ranking of general importance or authority in Web information retrieval. However, doing a PageRank computation for the whole Web graph is both time-consuming and costly. State of the art Web crawler based search engines also suffer from the latency in retrieving a complete Web graph for the computation of PageRank. We look into the problem of computing PageRank in a decentralized and timely fashion by making use of SiteRank and aggregating rankings from multiple sites. A SiteRank is basically the ranking generated by applying the classical PageRank algorithm to the graph of Web sites, i.e., the Web graph at the granularity of Web sites instead of Web pages. Our empirical results show that SiteRank also follows a power-law distribution. Our experimental results demonstrate that the decomposition of global Web document ranking computation by making use of SiteRank is a very promising approach for computing global document rankings in a decentralized Web search system. In particular, by sharing SiteRank among member servers, such a search system also obtains a new means to fight link spamming.
Keywords: Web information retrieval; link structure analysis; search engine; ranking algorithm; decentralized framework

Short Papers

Web Information Retrieval Based on User Profile BIBAFull-Text 275-278
  Rachid Arezki; Pascal Poncelet; Gérard Dray; David W. Pearson
With the growing popularity of the World Wide Web, the amount of available information is so great that finding the right and useful information becomes a very hard task for an end user. In this paper, we propose a new approach for personal Web information retrieval. The originality of our approach is a choice of indexing terms depending on the user request but also on his profile. The general idea is to consider that the need of a user depends on his request but also on his knowledge acquired through time on the thematic of his request.
Adaptive Support for Collaborative and Individual Learning (ASCIL): Integrating AHA! and CLAROLINE BIBAFull-Text 279-282
  Carlos Arteaga; Ramon Fabregat; Jorge Eyzaguirre; David Mérida
In this work we present a tool for Adaptive Support for Collaborative and Individual Learning (ASCIL). Taking as a base the fact that learning is not solely an individual nor collaborative process, we have come to the conclusion that it is an integrated activity which requires integral support. We have therefore applied two previously developed systems, AHA! and CLAROLINE, which we have integrated into ASCIL, a system which has the capacity to deliver adaptive support to individuals as well as collaborative learning. The interesting aspect of this present proposal is that the adaptive support for collaborative learning is integrated with the information contained in the User Model (student), which is kept in operation by AHA!. That is to say that the adaptive tasks are the result of individual learning in AHA! as a starting point.
Specification of Adaptive Behavior Using a General-Purpose Design Methodology for Dynamic Web Applications BIBAFull-Text 283-286
  Peter Barna; Geert-Jan Houben; Flavius Frasincar
Methodologies for the design and engineering of web applications evolve to accommodate the increased dynamic nature of modern web applications. In this paper we show and demonstrate the similarity between the dynamics in web applications and adaptive hypermedia systems using a general purpose model-driven web design methodology (Hera). To do so we use a simple example. We also stress advantages of specifying adaptivity within models defined on the schema level.
Using the X3D Language for Adaptive Manipulation of 3D Web Content BIBAFull-Text 287-290
  Luca Chittaro; Roberto Ranon
Web sites that include 3D content, i.e. Web sites where users navigate and interact (at least partially) through a 3D graphical interface, are increasingly employed in different domains, such as tutoring and training, tourism, e-commerce and scientific visualization. However, while a substantial body of literature and software tools is available about making 2D Web sites adaptive, very little has been published on the problem of personalizing 3D Web content and interaction. In this paper, we describe how we are exploiting a recently proposed 3D Web technology, i.e. the X3D (eXtensible 3D) language, for adaptive manipulation of 3D Web content.
Evaluation of APeLS -- An Adaptive eLearning Service Based on the Multi-model, Metadata-Driven Approach BIBAFull-Text 291-295
  Owen Conlan; Vincent P. Wade
The evaluation of learner and tutor feedback is essential in the production of high quality personalized eLearning services. There are few evaluations available in the Adaptive Hypermedia domain relative to the amount of research interest this domain is attracting. Many of the papers in this domain focus on the technological design of systems without justifying the designs through the lessons learned from evaluations. This paper evaluates the usability and effectiveness of using the multi-model, metadata-driven approach for producing rich adaptive eLearning solutions that remain content and domain independent. Through this independence, the eLearning services developed can utilize many pedagogical approaches and a variety of models to produce a wide range of highly flexible solutions. This paper identifies benefits to learners brought through adopting the multi-model approach gathered over four years of student evaluation. It briefly describes the evaluation of the Adaptive Personalized eLearning Service (APeLS), a personalized eLearning service based on a generic adaptive engine.
SearchGuide: Beyond the Results Page BIBAFull-Text 296-299
  Maurice Coyle; Barry Smyth
Today's Web users are frequently frustrated at their inability to efficiently locate specific items of interest on the Internet. This is mainly due to the sheer size and speed of growth of the Web; recent estimates suggest it contains more than 10 billion pages and that it is growing by 60 terabytes per day [1]. Web search engines are the primary way that users hunt for information but we argue that in their present form they do not go far enough to help users locate relevant information. We investigate ways of aiding the user past the initial results page by leveraging information from previous search sessions.
Modeling Learners as Individuals and as Groups BIBAFull-Text 300-303
  Roland Hübscher; Sadhana Puntambekar
Adaptive navigation support normally attempts to make selecting a relevant hyperlink as easy as possible. However, in educational applications, this may have negative learning effects since selecting a link is sometimes an important educational problem for the student to solve. To provide appropriate scaffolding to students, it is necessary to understand how they navigate in hypermedia sites. By grouping students with similar conceptual (mis)understanding we were able to uncover a small set of characteristic navigation patterns, and to demonstrate that students with similar conceptual understanding have similar navigation patterns.
Adaptive Help for Webbased Applications BIBAFull-Text 304-307
  Dorothea Iglezakis
This paper presents an approach that uses the techniques of plan recognition not only to infer short-term plans and goals, but also to infer the long-term procedural knowledge of a user in a non-binary way. The information about the procedural knowledge in terms of activation builds the user model of AdaptHelp, an adaptive help system for web-based systems. AdaptHelp is based onto established adaptive help systems, which are shortly presented and compared on the used mechanism and techniques.
Empirical Evaluation of an Adaptive Multiple Intelligence Based Tutoring System BIBAFull-Text 308-311
  Declan Kelly; Brendan Tangney
EDUCE is an Intelligent Tutoring System for which a set of learning resources has been developed using the principles of Multiple Intelligences. It can dynamically identify learning characteristics and adaptively provide a customised learning material tailored to the learner. This paper describes a research study using EDUCE that examines the relationship between the adaptive presentation strategy, the level of choice available and the learning performance of science school students aged 12 to 14. The paper presents some preliminary results from a group of 18 students that have participated in the study so far. Results suggest that learning strategies that encourage the student to use as many resources as possible are the most effective. They suggest that learning gain can improve by presenting students initially with learning resources that are not usually used and subsequently providing a range of resources from which students may choose.
Evaluating Information Filtering Techniques in an Adaptive Recommender System BIBAFull-Text 312-315
  John O'Donovan; John Dunnion
With the huge increase in the volume of information available in digital form and the increasing diversity of Web applications, the need for efficient, reliable information filtering is critical. New algorithms that filter information for specific tastes are being developed to tackle the problem of information overload. This paper proposes that there is a substantial relative difference in the performances of various filtering algorithms as they are applied to different datasets, and that these performance differences can be leveraged to form the basis of an Adaptive Information Filtering System. We classify five different datasets based on a number of metrics, including sparsity, ratings distribution and user-item ratio, and develop a regression function over these metrics to predict the suitability of a particular recommendation algorithm for a previously unseen dataset. Our results show that the predicted best algorithm does perform best on the new dataset.
Adaptive Educational Hypermedia Proposal Based on Learning Styles and Quality Evaluation BIBAFull-Text 316-319
  Marcela Prieto Ferraro; Helmut Leighton Álvarez; Francisco García Peñalvo
This is a proposal of how to determine quality attributes in the Adaptive Educational Hypermedia Systems based on Learning Styles for a later design of a quality evaluation methodology of these systems. Some specific learning styles and their relationships with the outlined instructional strategies are examined, in order to find in a further work a way of how to determine the quality attributes and standards for the elaboration of the quality evaluation methodology for this systems.
Adaptive Course Player for Individual Learning Styles BIBAFull-Text 320-323
  Katja Reinhardt; Stefan Apelt; Marcus Specht
The paper describes the development and implementation of an adaptive course player that uses standardized learning materials, metadata, and a learning style model based on the Felder-Silverman learning style classification. The system implements adaptation of individual recommendations and content adaptation based on learning styles.
Rhetorical Patterns for Adaptive Video Documentaries BIBAFull-Text 324-327
  Cesare Rocchi; Massimo Zancanaro
In this paper, we introduce an approach to the adaptive composition of video documentaries. The adaptation is based on templates that encode rules for the dynamic selection, sequencing and composition of video shots. We introduce a template language to define adaptation rules. Finally, we discuss rhetorical patterns, strategies that we abstracted out during the realization of a museum mobile guide.
Location-Aware Adaptive Interfaces for Information Access with Handheld Computers BIBAFull-Text 328-331
  Golha Sharifi; Ralph Deters; Julita Vassileva; Susan Bull; Harald Röbig
Adapting to user context versus adapting to individual user features or behaviour patterns has been a topic of recent discussion. We believe both types of adaptation are valuable and the decision of which to apply, or how to combine the two, is domain or application specific. As an illustration, this paper presents two approaches to adapting the interface according to the type of user and the extent to which the user's task and location is predetermined.
PSO: A Language for Web Information Extraction and Web Page Clipping BIBAFull-Text 332-335
  Tetsuya Suzuki; Takehiro Tokuda
Web information extraction and Web page clipping are important technique in adaptive hypermedia which adapt or compose contents from real Web pages. A technical problem for them is robustness of specification of parts of Web pages against update of Web pages. In this paper we present a language called PSO for Web information extraction and Web page clipping. Thanks to operations called path set operations which the language provides for the problem, we can specify parts of Web pages with small differences in expected formats by structural similarity between Web pages. We also show application of the language to real Web pages.
Swarm-Based Adaptation: Wayfinding Support for Lifelong Learners BIBAFull-Text 336-339
  Colin Tattersall; Bert van den Berg; René van Es; José Janssen; Jocelyn Manderveld
This article introduces an approach to adaptive wayfinding support for lifelong learners based on self-organisation theory. It describes an architecture which supports the recording, processing and presentation of collective learner behaviour designed to create a feedback loop informing learners of successful paths towards the attainment of their learning objectives. The approach is presented as an alternative to methods of achieving adaptation in hypermedia-based learning environments which involve learner modelling.
Giving More Adaptation Flexibility to Authors of Adaptive Assessments BIBAFull-Text 340-343
  Aimilia Tzanavari; Symeon Retalis; Panikos Pastellis
In this paper, we present AthenaQTI, a tool for authoring personalized assessments, which gives the author significant flexibility in terms of the adaptation that s/he can incorporate in the assessments s/he builds. We focus on presenting the functionality of the authoring environment and the tool's conformance to the IMS- QTI specification, a fact that gives it the advantage of interoperability. Furthermore, we briefly describe the user model and the philosophy of its manipulation.
A Generic Adaptivity Model in Adaptive Hypermedia BIBAFull-Text 344-347
  Paul de Vrieze; Patrick van Bommel; Theo van der Weide
For adaptive hypermedia there is a strong model in form of the AHAM model and the AHA! system. This model, based on the Dexter Model, however is limited to application in hypermedia systems. In this paper we propose a new Generic Adaptivity Model. This state-machine based model can be used as the basis for adaptation in all kinds of applications.

Doctoral Consortium

Extreme Adaptivity BIBAFull-Text 348-352
  Mário Amado Alves; Alípio Jorge; José Paulo Leal
This Doctoral Consortium paper focuses on Extreme Adaptivity, a set of top level requirements for adaptive hypertext systems, which has resulted from one year of examining the adaptive hypertext landscape. The complete specification of a system, KnowledgeAtoms, is also given, mainly as an example of Extreme Adaptivity. Additional methodological elements are discussed.
A Learner Model in a Distributed Environment BIBAFull-Text 353-359
  Cristina Carmona; Ricardo Conejo
A learner model must store all the relevant information about a student, including knowledge and attitude. This paper proposes a domain independent learner model based in the classical overlay approach that can be used in a distributed environment. The model has two sub-models: the learner attitude model, where the static information about the user is stored (user's personal and technical characteristics, user's preferences, etc.) and the learner knowledge model, where the user's knowledge and performance is stored. The knowledge model has four layers: estimated, assessed, inferred by prerequisite and inferred by granularity. The learner model is used as a part of the MEDEA system, so the first and second layers are updated directly by the components of MEDEA and the third and fourth are updated by Bayesian inference.
A Semantic Meta-model for Adaptive Hypermedia Systems BIBAFull-Text 360-365
  Patricia Seefelder de Assis; Daniel Schwabe
In this work we show a general meta-model for Adaptive Hypermedia Systems (AHSs) and discuss which research issues are necessary in order to improve this model. We argue that this meta-model shall then be described using ontologies from the Semantic Web and argument that a Semantic Meta-Model for AHSs may improve adaptation and serve as a basis for meta-adaptation.
Adaptive Navigation for Self-assessment Quizzes BIBAFull-Text 366-371
  Sergey Sosnovsky
Web-based parameterized quizzes provide teachers and students with several advantages as the technology for self-assessment. However, the effect of these advantages is strongly reduced, if a student does not receive enough support and cannot see her/his progress during the course. We have developed the system QuizGuide, which attempts to solve this problem by navigating students through the quiz material of the course in adaptive way. Architecture, interface and plans of future system development are described here. The paper also presents first results of system evaluation.

Posters

Towards Adaptive Learning Designs BIBAFull-Text 372-375
  Adriana Berlanga; Francisco J. García
This paper outlines an ongoing research. Its objective is to propose an open model to define adaptive learning designs in such way that novice designers or teachers could configure them. Further, all elements are annotated with standardized metadata to facilitate their exchangeability and reusability.
Time-Based Extensions to Adaptation Techniques BIBAFull-Text 376-379
  Mária Bieliková; Rastislav Habala
In this paper we present an approach to time-based adaptation in adaptive hypermedia systems. Time is used as a part of the context (or environment) model. We have proposed extensions of known adaptation techniques by means of the notion of time. We experimented with proposed extensions and implemented a software system called TIM, which adapts presentation of educational module administrative information according user characteristics and a context represented by time of information presentation.
On the Use of Collaborative Filtering Techniques for the Prediction of Web Search Result Rank BIBAFull-Text 380-383
  Peter Briggs; Barry Smyth
In this paper we describe an experiment that applies collaborative filtering techniques to a Web search task, namely the prediction of the rank of relevant results for a given query. We compare the performance of two different collaborative filtering algorithms and argue that these approaches have the potential to generate accurate predictions, which in turn may help to improve existing Web search techniques.
A Thematic Guided Tour Model for Contextualized Concept Presentations BIBAFull-Text 384-388
  Benjamin Buffereau; Philippe Picouet
Since adaptive hypermedia systems generally confront with the difficulty to combine concept-based hyperspace generation with multi-concept indexing, we propose in this article an adaptive hypermedia model that achieves a fruitful separation of the resource space and the user space. On the one hand, the resource space benefits a powerful indexing model. On the other hand, the user space is generated from the knowledge space as thematic guided tours specified at a schema level. These guided tours take advantage of the indexing model to contextualize concept presentations and ensure global and local coherence. The model fulfills the requirements of an open, time-evolving, multi-author AHS and can be applied to a community web context.
A Fuzzy Set Based Tutoring System for Adaptive Learning BIBAFull-Text 389-392
  Sook-young Choi; Hyung-jung Yang
This paper proposes a web-based adaptive tutoring system based on fuzzy set that provides learning materials and questions dynamically according to students' knowledge state and gives advices for the learning after an evaluation. For this, we construct a fuzzy level set considering the importance degree of learning goals, the difficulty degree of learning materials, and the relation degree between learning goals and learning materials. Using the fuzzy level set, our system offers learning materials and questions adapted to individual students. Moreover, a result of the test is evaluated with fuzzy linguistic variable.
Offering Collaborative-Like Recommendations When Data Is Sparse: The Case of Attraction-Weighted Information Filtering BIBAFull-Text 393-396
  Arnaud De Bruyn; C. Lee Giles; David M. Pennock
We propose a low-dimensional weighting scheme to map information filtering recommendations into more relevant, collaborative filtering-like recommendations. Similarly to content-based systems, the closest (most similar) items are recommended, but distances between items are weighted by attraction indexes representing existing customers' preferences. Hence, the most preferred items are closer to all the other points in the space, and consequently more likely to be recommended. The approach is especially suitable when data is sparse, since attraction weights need only be computed across items, rather than for all user-item pairs. A first study conducted with consumers within an online bookseller context, indicates that our approach has merits: recommendations made by our attraction-weighted information filtering recommender system significantly outperform pure information filtering recommendations, and favorably compare to data-hungry collaborative filtering systems.
Using Concept Maps for Enhancing Adaptation Processes in Declarative Knowledge Learning BIBAFull-Text 397-400
  Fabien Delorme; Nicolas Delestre; Jean-Pierre Pécuchet
This article deals with adaptation techniques in the field of declarative knowledge learning. After explaining how concepts can be represented, it introduces a learner evaluation technique based on a concept maps analysis. The way an epistemic learner model can be made from this evaluation is then proposed. Finally, adaptation techniques based on this model are presented. With this method, different adaptation schemes can be applied to the document depending on the learner's errors.
An Adaptive Tutoring System Based on Hierarchical Graphs BIBAFull-Text 401-404
  Sergio Gutiérrez; Abelardo Pardo; Carlos Delgado Kloos
An adaptive tutoring system is presented based on hierarchical graphs that capture the sequencing of a set of learning objects depending on how students interact with them. The use of hierarchy allows the definition of complex transition structures over arbitrarily large sets of objects. Using this approach a tutoring tool has been designed and tested in the context of an introductory course in Computer Architecture. Experimental results clearly show the positive impact of the proposed content adaptation over how students learn concepts.
A Brief Introduction to the New Architecture of SIETTE BIBAFull-Text 405-408
  Eduardo Guzmán; Ricardo Conejo
SIETTE is a web-based adaptive testing system released some years ago. It implements Computerized Adaptive Tests. In these tests the selection of the questions posed to students, the decision to finalize the test is accomplished adaptively. In this paper, we present the new architecture of SIETTE and some new features recently implemented.
A Reinforcement Learning Approach to Achieve Unobtrusive and Interactive Recommendation Systems for Web-Based Communities BIBAFull-Text 409-412
  Felix Hernandez; Elena Gaudioso; Jesús G. Boticario
Adaptive recommendation systems build a list of suggested links to nodes that usually cannot be reached directly from current web page. These recommendations are given by means of user models, where some parts of those models may be mined/learned from user's interactions with a web site.
   However, user's interactions with the web site do not usually include user's interaction with the recommendation system. In other words, most of current systems adapt recommendations to users just by "looking over her shoulder". That occurs, in spite of the fact that taking into account user's behavior upon recommendation should be a main part of the adaptation mechanism, because recommendation is not transparent to a user.
   Other recommendation systems interact with the user, but in an obtrusive way, making explicit requests to the user (prompting the user for rating) that are usually not followed. In this paper we present a recommendation system "in front of the user", a system that looks directly at the user and interacts with her softly. Its key features are (i) it adapts to users by taking into account their interactions with a Web-based Communities Platform, and (ii) it adapts its own recommendations by unobtrusively taking into account the user behavior upon recommendations.
GEAHS: A Generic Educational Adaptive Hypermedia System Based on Situation Calculus BIBAFull-Text 413-416
  Cédric Jacquiot; Yolaine Bourda; Fabrice Popineau
GEAHS is a platform designed to ease the development of Adaptive Educational Hypermedia, using standard formalisms. In this document, we explain the underlying principles of this platform. Genericity is achieved thanks to an adaptation engine based on situation calculus and RDF. This paper describes the main aspects of our system, as well as the use we make of situation calculus to create a simpler, more reusable adaptive hypermedia system.
Problem Solving with Adaptive Feedback BIBAFull-Text 417-420
  Rainer Lütticke
The virtual laboratory (VILAB) supports interactive problem solving in computer science with access to complex software-tools. During the problem solving processes the learners get fast feedback by a tutoring component. This feedback based in the first version of VILAB only on intelligent error analyses of learners' solutions. Animated by the very positive results of the evaluation of this tutoring component we additionally implemented an user model. Thereby the feedback for a learner consists not only of adaptive information about his errors and performance, but also of adaptive hints for the improvement of his solution. Furthermore, the tutoring component can individually motivate the learners.
Machine Learning Methods for One-Session Ahead Prediction of Accesses to Page Categories BIBAFull-Text 421-424
  José D. Martín-Guerrero; Emili Balaguer-Ballester; Gustavo Camps-Valls
This paper presents a comparison among several well-known machine learning techniques when they are used to carry out a one-session ahead prediction of page categories. We use records belonging to 18 different categories accessed by users on the citizen web portal Infoville XXI. Our first approach is focused on predicting the frequency of accesses (normalized to the unity) corresponding to the user's next session. We have utilized Associative Memories (AMs), Classification and Regression Trees (CARTs), Multilayer Perceptrons (MLPs), and Support Vector Machines (SVMs). The Success Ratio (SR) averaged over all services is higher than 80% using any of these techniques. Nevertheless, given the numerous quantity of services taken into account, and the variability of SR among them, a balanced performance is desirable. When this issue is analysed, SVMs yielded the best overall performance. This study suggests that a prediction engine can be useful in order to customize user's interface.
Gender-Biased Adaptations in Educational Adaptive Hypermedia BIBAFull-Text 425-428
  Erica Melis; Carsten Ullrich
Studies (for instance, [1,2]) show that there is a statistically relevant gender difference in computer usage. In this paper we address the question of what are the causes of this problem and how it can be relieved by adaptive means.
   In trying to design systems that are sensitive to individual differences there is a dilemma -- how can you make the system behave in a way appropriate for each individual without forcing people into stereotypes? We propose that the group characteristics can be taken as weak defaults and coupled with an adaptive mechanism to quickly take account of individual differences.
   This paper starts with a summary of a study which derives a model of the mental factors that influence computer usage. Then, we refine that model and make it the basis of a Bayesian Net student model. The main section describes suggestions for Adaptive Hypermedia targeting the relevant mental factors and how adaptivity can help to avoid clichés and thus discrimination.
An Overview of aLFanet: An Adaptive iLMS Based on Standards BIBAFull-Text 429-432
  Olga C. Santos; Carmen Barrera; Jesús G. Boticario
aLFanet (IST-2001-33288) aims to build an adaptive iLMS (intelligent Learning Management System) that provides personalised eLearning based on the combination of different types of adaptation (e.g. learning routes, interactions in services, peer-to-peer collaboration, presentation). It integrates new principles and tools in the fields of Learning Design and Artificial Intelligence, following existing standards in the educational field (IMS-LD, IMS-CP, IEEE-LOM, IMS-LIP, IMS-QTI) and multi-agents systems (FIPA). In this paper we present an overview of the project ongoing research and developments.
A General Meta-model for Adaptive Hypermedia Systems BIBAFull-Text 433-436
  Patricia Seefelder de Assis; Daniel Schwabe
The emergence of meta-adaptive systems seems to be a natural evolution process in adaptive systems. Meta-adaptive systems are able to vary the adaptation technique based on various factors. This work analyzes the main aspects related to adaptivity by using a general meta-model of Adaptive Hypermedia Systems, intended to serve as a basis for a meta-adaptive system.
Adaptive Services for Customised Knowledge Delivery BIBAFull-Text 437-440
  Alexander Smirnov; Mikhail Pashkin; Nikolai Chilov; Tatiana Levashova
The more information is available in the Internet the more difficult to find the right information. The idea of adaptive hypermedia systems is to provide users with personalised information that depends on the users' needs and requirements. The paper presents an approach to knowledge logistics that applies such principles to the process of knowledge delivery from distributed heterogeneous sources. More attention is paid to description of an adaptive service that can modify itself based on the user tasks and implemented within the framework of the approach.