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

Proceedings of AH 2000 Adaptive Hypermedia and Adaptive Web-based Systems 2000-08-28

Fullname:Proceedings of the International Conference on Adaptive Hypermedia and Adaptive Web-Based Systems
Editors:Peter Brusilovsky; Oliviero Stock; Carlo Strapparava
Location:Trento, Italy
Dates:2000-Aug-28 to 2000-Aug-30
Series:Lecture Notes in Computer Science, 2000, Volume 1892
Standard No:ISBN: 978-3-540-67910-3 (Print) 978-3-540-44595-1 (Online); hcibib: UMAP00
Links:Online Proceedings | Conference Home Page
Summary:Web-based application systems, as well as other complex hypermedia systems with a large variety of users, suffer from an inability to satisfy heterogeneous needs. A Web course presents the same static explanation of a concept to students with widely differing knowledge of the subject. A Web bookstore offers the same selection of bestsellers to customers with different reading preferences. A Web museum offers the same "guided tour" and the same narration to visitors with very different goals and interests. A remedy for the negative effects of the traditional "one-size-fits-all" approach is to enhance a system's ability to adapt its own behavior to the goals, tasks, interests, and other features of individual users. Starting in the 1990s, many research teams began to investigate ways of modeling features of the users of hypermedia systems. This has led to a number of interesting adaptation techniques and adaptive hypermedia systems. The Web, with its clear demand for personalization, served as a real booster for this research area, providing both a challenge and an attractive platform.
    The International Conference on Adaptive Hypermedia and Adaptive Webbased Systems has continued and joined together two successful workshop series: the Workshops on Adaptive Hypertext and Hypermedia and the Workshops on Adaptive Systems and User Modeling on the World Wide Web previously held in conjunction with such international conferences as User Modeling, ACM Hypertext, and World Wide Web Conference. These workshops were so wellreceived by the international community that the organizers decided to proceed with a separate conference in the year 2000.
    Due to its interdisciplinary nature, the conference has attracted a large number of submissions from researchers with very different backgrounds such as hypertext, user modeling, machine learning, natural language generation, information retrieval, intelligent tutoring systems, cognitive science, and Web-based education. Continuing the tradition of earlier workshops, AH 2000 provided a forum in which researchers and practitioners with different backgrounds could exchange their complementary insights. Overall AH 2000 embodied 4 invited talks, 22 full-paper presentations (selected from 55 submitted), 31 short-paper presentations, and 4 presentations at the Doctoral Consortium. With the exception of some of the invited talks, all presented papers can be found in these proceedings.
    The logistics involved in organizing the first full conference of this kind were not trivial. The help from many people and organizations was important to make the conference and the proceedings reach fruition. ITC-irst was glad to host the first European attempt to put together researchers of this field in a conference devoted to this topic. The European Commission sponsorship was very important. We gratefully acknowledge it and consider it a sign of the strategic relevance of this theme. We thank the Program Committee members and the external reviewers for their excellent job in reviewing the unexpectedly large number of submissions. We gratefully acknowledge the help from AH 2000 cooperative societies - AI*IA Associazione Italiana per l'Intelligenza Artificiale, Association for Computing Machinery and its Special Interest Groups SIGART, SIGCHI, SIGIR, and SIGWEB, International Artificial Intelligence in Education Society, and User Modeling Inc. All of them have helped us to deliver the information about AH 2000 to a large number of researchers worldwide. Finally, we are thankful to IJCAI for providing a "conference seeding grant" that has enabled a number of students to attend AH 2000 and to Kluwer Academic Publishers (the publisher of User Modeling and User-Adapted Interaction) for supporting the best paper award.
  1. Invited Paper
  2. Full Papers
  3. Short Papers
  4. Doctoral Consortium Papers

Invited Paper

Enhancing Adaptive Hypermedia Presentation Systems 1 by Lifelike Synthetic Characters BIBAFull-Text 1-4
  Elisabeth André
Rapid growth of competition in the electronic market place will boost the demand for new innovative communication styles to attract web users. With the advent of web browsers that are able to execute programs embedded in web pages, the use of animated characters for the presentation of information over the web has become possible. Instead of surfing the web on their own, users can join a tour, ask the lifelike character for assistance or even delegate a complex search task to it.

Full Papers

Dynamic Generation of Adaptive Web Catalogs BIBAFull-Text 5-16
  Liliana Ardissono; Anna Goy
This paper describes the techniques used to dynamically generate personalized Web catalog pages in a prototype toolkit for the creation of adaptive Web stores. We focus on the integration of personalization strategies for selecting the layout and content of the catalog pages, with Natural Language Generation techniques, used to dynamically produce the descriptions of products, tailored to the individual user.
An Intelligent Tutor for a Web-Based Chess Course BIBAFull-Text 17-26
  Antonio Baena; María-Victoria Belmonte; Lawrence Mandow
Web-based intelligent tutoring systems try to fill the gap between human teachers and printed textbooks as distance learning aids. Actually, intelligent tutoring systems research is concerned with the development of computer tools that show adaptive capabilities in the domain of tutoring, where the student's progress is autonomously monitored and guided according to some tutoring strategy. This paper provides details on the analysis, design and implementation of such a system. STIA (Sistema Tutor en Internet de Ajedrez) is a fully implemented Web-based tool developed to provide adaptive guidance and help while learning chess basics. In STIA the task of the tutor is to guide the student efficiently through the course material, according to the tutoring strategy defined by the course authors. This is achieved in two ways. First, it imposes limits on the portion of course material the students can access. This prevents them from getting lost in a flood of information. Second, the tutor evaluates each student's knowledge through a set of problems, and according to result recommends reviewing theory, solving more problems or advancing through the course.
Adapting Web-Based Information to the Needs of Patients with Cancer BIBAFull-Text 27-37
  Diana Bental; Alison Cawsey; Janne Pearson; Ray Jones
Good patient education can help to reduce health service costs and improve the quality of life of people with chronic or terminal conditions. Adapting educational materials to the patients' needs and interests can make them more effective. Computer-based techniques make this adaptation more feasible.
   In this paper we describe a theoretically motivated framework for the provision of computer-based information for cancer patients, and the computational techniques used to implement it. Our goal is to develop an interactive hypertext system which could provide patients with the right information at the right time, avoiding some of the need to search through the copious literature available.
   The paper describes how we use an explicit model of relevance to select and present information that is adapted at different levels to the situational and process-based aspects of the patient's illness and treatment.
Group User Models for Personalized Hyperlink Recommendations BIBAFull-Text 38-50
  Johan Bollen
This paper presents a system that combines adaptive hypertext linking based on group link preferences with an implicit navigation-based mechanism for personalized link recommendations. A methodology using three Hebbian-style learning rules changes hyperlink weights according to users' overlapping navigation paths and causes a hypertext system's link structure to converge to a valid group user model. A spreading activation recommendation system generates navigation path based recommendations for individual users. Both systems are linked, thereby combining both personal user interests and established group link preferences. An on-line application for the Los Alamos National Laboratory Research Library is presented.
Adaptive Navigation Support and Adaptive Collaboration Support in WebDL BIBAFull-Text 51-61
  Jesus G. Boticario; Elena Gaudioso; Felix Hernandez
This article presents adaptive navigation support and adaptive collaboration support tasks in WebDL, an interactive system for focusing teaching on student performance and resolving problems detected in Internet use for distance learning. This adaptation is done through user model acquisition from the student data available and interaction with the system. WebDL combines techniques used in intelligent tutoring systems, adaptive hypermedia programs and learning apprentice systems for software personalization.
Case-Based User Profiling for Content Personalisation BIBAFull-Text 62-72
  Keith Bradley; Rachael Rafter; Barry Smyth
As it stands the Internet's "one size fits all" approach to information retrieval presents the average user with a serious information overload problem. Adaptive hypermedia systems can provide a solution to this problem by learning about the implicit and explicit preferences of individual users and using this information to personalise information retrieval processes. We describe and evaluate a two-stage personalised information retrieval system that combines a server-side similarity-based retrieval component with a client-side case-based personalisation component. We argue that this combination has a number of benefits in terms of personalisation accuracy, computational cost, flexibility, security and privacy.
Providing Tailored (Context-Aware) Information to City Visitors BIBAFull-Text 73-85
  Keith Cheverst; Nigel Davies; Keith Mitchell; Paul Smith
The GUIDE system has been developed in order to provide city visitors with an intelligent and context-aware tourist guide. The system has been deployed in the city of Lancaster and integrates the use of hand-held personal computing technologies, wireless communications, context-awareness and adaptive hypermedia. This paper focuses on the role of adaptive hypermedia within the GUIDE system and the techniques used to tailor or adapt the presentation of web-based information. The context used by GUIDE includes the visitor's personal context, e.g. the visitor's current location and personal profile, and the environmental context, e.g. the opening times of the city's attractions. Following a field trial based evaluation, in which positive feedback was received, the system is now publicly available to visitors who wish to explore the city.
Adding Adaptive Features to Virtual Reality Interfaces for E-Commerce BIBAFull-Text 86-97
  Luca Chittaro; Roberto Ration
Virtual Reality (VR) interfaces to e-commerce sites have recently begun to appear on the Internet, promising to make the e-shopping experience more natural, attractive, and fun for customers. Adaptivity is an important issue for these VR applications, because it would make them suitable for the 1-to-1 e-commerce strategies towards which sellers are increasingly driven. It is thus surprising that the introduction of adaptive features in VR stores remains a completely unexplored issue. This paper begins to face the problem, presenting and discussing ADVIRT, a first prototype of an adaptive VR store. In ADVIRT, a set of personalization rules exploits a model of the customer to adapt features of the VR store such as: (i) the display of different products in the store (e.g., shelf space, display spots, banners, audio advertising), (ii) the navigation aids available to the customer, (iii) the store layout, organization, and look.
WAPing the Web: Content Personalisation for WAP-Enabled Devices BIBAFull-Text 98-108
  Paul Cotter; Barry Smyth
Content personalisation technologies may hold the key to solving the information overload problem associated with the Internet, by facilitating the development of information services that are customised for the needs of individual users. For example, PTV is an award-winning, Web-based personalised television listings service capable of learning about the viewing habits of individual users and of generating personalised TV guides for these users. This paper describes how PTV has been recently adapted for use on the new generation of WAP-enabled Internet devices such as mobile phones -- the need for content personalisation is even more acute on WAP devices due to their restricted presentation capabilities.
Extendible Adaptive Hypermedia Courseware: Integrating Different Courses and Web Material BIBAKFull-Text 109-120
  Nicola Henze; Wolfgang Nejdl
Adaptive hypermedia courseware benefits from being distributed over the Web: content can always be kept up-to-date, discussions and interactions between instructors and learners can be supported, new courses can easily be distributed to the students. Nevertheless, adaptive hypermedia systems are -- even in the web content -- still stand-alone systems as long as they lack the ability to integrate and adapt information from arbitrary places in the web.
   In this paper, we discuss the integration of hypermedia courses and web material into existing, adaptive hypermedia courses. We show a possible solution which we have used for an undergraduate course about Java programming. We then discuss this solution as well as advantages and problems, and identify several research issues which have still to be solved for adapting distributed course materials.
Keywords: Open adaptive hypermedia systems; educational hypermedia systems
Logically Optimal Curriculum Sequences for Adaptive Hypermedia Systems BIBAFull-Text 121-132
  Roland Hübscher
Curriculum sequencing is an important technique used in many adaptive hypermedia systems. When following one of the possible page sequences, visiting some pages may become redundant, because its content has been covered already by another page. Using disjunctive and conjunctive prerequisites instead of partial orders to describe the many possible sequences, logical redundancy between pages can be computed on the fly without burdening the teaching model with that task [1]. Although the general case of finding all redundant pages is NP-Complete [2] and thus, intractable unless P = NP, a large subset can be located efficiently in realtime. The advantage of separating out logical redundancy, the advantage of using conjunctive and disjunctive prerequisites, and the algorithms to find redundant pages are discussed. An interesting characteristic of the presented approach is that it can be used together with a wide variety of user and teaching models.
Towards Zero-Input Personalization: Referrer-Based Page Prediction BIBAFull-Text 133-143
  Nicholas Kushmerick; James McKee; Fergus Toolan
Most web services take a "one size fits all" approach: all visitors see the same generic content, formatted in the same generic manner. But of course each visitor has her own information needs and preferences. In contrast to most personalization systems, we are interested in how effective personalization can be with zero additional user input or feedback. This paper describes PWW, an extensible suite of tools for personalizing web sites, and introduces RBPR, a novel zero-input recommendation technique. RBPR uses information about a visitor's browsing context (specifically, the referrer URL provided by HTTP) to suggest pages that might be relevant to the visitor's underlying information need. Empirical results for an actual web site demonstrate that RBPR makes useful suggestions even though it places no additional burden on web visitors.
LiveInfo: Adapting Web Experience by Customization and Annotation BIBAFull-Text 144-154
  Paul P. Maglio; Stephen Farrell
Intermediaries are perfectly suited to customizing and annotating web pages, as they can stand in the flow of data between web browser and web server, monitoring user behavior and modifying page markup. In this paper, we present LiveInfo, an intermediary-based framework for customizing and annotating web pages. LiveInfo breaks customization and annotation into four steps: (a) splitting streaming data into useful chunks, (b) identifying meaningful patterns of chunks, (c) merging together overlapping patterns, and (d) adding markup to customize and annotate. Each of these steps is easily replaced or configured, making it simple to adapt web experience by customization and annotation.
Adaptivity for Conceptual and Narrative Flow in Hyperbooks: The MetaLinks System BIBAFull-Text 155-166
  Tom Murray; Tina Shen; Janette Piemonte; Chris Condit; Jason Thibedeau
In this paper we discuss MetaLinks, a framework and authoring tool for web-based adaptive hyper-books. We focus on how features of the system address the problem issues of disorientation, cognitive overload, discontinuous flow (poor narrative flow or poor conceptual flow), and content non-readiness.
The MacroNode Approach: Mediating Between Adaptive and Dynamic Hypermedia BIBAFull-Text 167-178
  Elena Not; Massimo Zancanaro
In this paper, we discuss an approach that tries to blur the distinction between adaptive hypermedia and dynamic NLG-based hypermedia. The approach aims at finding an optimal trade-off between resource reuse and flexibility: existing atomic pieces of data are collected and properly annotated; at the interaction time, the system dynamically builds the nodes of the hypermedia composing different pieces together. The proposed annotation formalism is illustrated and a rule-based system to compose hypermedia nodes exploiting different knowledge sources is presented. Finally, the advantages of this approach with respect to adaptation and dynamic generation are discussed.
ECHOES: An Immersive Training Experience BIBAFull-Text 179-188
  Gregory O'Hare; Katherine Sewell; Aidan Murphy; Thomas Delahunty
The main objective of the ECHOES[1] project is to build a distributed, adaptive, dynamic environment for educating and supporting technicians in the use and maintenance of complex industrial artefacts. To pursue this objective, Web Based Training, Virtual Reality and Multi-Agent Systems are synthesised within the ECHOES environment. Users co-exist within the virtual environment and with time social cohesion emerges yielding a virtual community. User profiling techniques facilitate adaptive courseware presentation.
A Connectionist Approach for Supporting Personalized Learning in a Web-Based Learning Environment BIBAFull-Text 189-201
  Kyparisia A. Papanikolaou; George D. Magoulas; Maria Grigoriadou
The paper investigates the use of computational intelligence for adaptive lesson presentation in a Web-based learning environment. A specialized connectionist architecture is developed and a formulation of the planning strategy retrieval in the context of the network dynamics is proposed to select the content of the lesson in a goal-oriented way of 'teaching'. The educational material of the course is stored in a connectionist-based distributed information storage system that provides capabilities for optimal selection of the educational material according to the knowledge needs, abilities and preferences of each learner. Low-level tests of the system have been performed to investigate how the connectionist architecture and the learner model function together to create an operational learning environment. Preliminary experiments indicate that personalized content delivery is provided in an educational effective way.
Adaptive Hypertext Design Environments: Putting Principles into Practice BIBAFull-Text 202-213
  Daniela Petrelli; Daniele Baggio; Giovanni Pezzulo
This paper discusses the design of a tool for authoring adaptive hypertext. First we describe the task of adaptive hypertext design. The network editing task as well as the testing phase are explored showing the importance of a design environment for authors and developers of hypertext adaptive systems. Then the required support such environment has to provide are presented. Finally we describe how principles can be translated into practice by illustrating the implementation of a design environment for developing and testing a very flexible adaptive hypertext.
ECSAIWeb: A Web-Based Authoring System to Create Adaptive Learning Systems BIBAFull-Text 214-226
  Charun Sanrach; Monique Grandbastien
ECSAIWeb is a variant of ECSAI, an environment for designing intelligent tutoring systems. The tutoring knowledge contained in a tutor agent designed to run in a non-networked environment is reused and updated for the World Wide Web. We concentrate on a flexible architecture that allows teachers or authors to modify and add their knowledge in the domain. We apply adaptive techniques: adaptive presentations and adaptive navigation to present the domain knowledge to learners.
Adaptive Content in an Online Lecture System BIBAFull-Text 227-238
  Mia K. Stern; Beverly Park Woolf
This paper discusses techniques for adapting the content in an online lecture system for a specific user. A two pass method is used: 1) determine the appropriate level of difficulty for the student and 2) consider the student's learning style preferences. A simple grading scheme is used to determine the student's knowledge and a Naïve Bayes Classifier is used to reason about the student's preferences in terms of explanations, examples, and graphics. A technique for gathering and using population data is also discussed.
A Web-Based Socratic Tutor for Trees Recognition BIBAFull-Text 239-249
  Mónica Trella; Ricardo Conejo; Eduardo Guzman
Socratic dialogues has been widely uses as a way of implement an ITS. The idea behind it is that the teaching and learning process should be based upon a personal reflection that can be obtained posing the right question on a guided dialogue. This methodology assumes that the knowledge acquisition is a discovering process in which both the teacher and the student plays an active role. This tutorial strategy has been developed as a part of a web based ITS architecture for declarative domains and it has been applied to the botanical domain. In this paper we describe this component, the knowledge representation that support it and the web interface used.
Adaptation Control in Adaptive Hypermedia Systems BIBAKFull-Text 250-259
  Hongjing Wu; Paul De Bra; Ad Aerts; Geert-Jan Houben
A hypermedia application offers its users a lot of freedom to navigate through a large hyperspace, described by a domain model. Adaptive hypermedia systems (AHS) aim at overcoming possible navigation and comprehension problems by providing adaptive navigation support and adaptive content. The adaptation is based on a user model that represents relevant aspects about the user. In this paper, we concentrate on the adaptation engine (AE) that is responsible for performing the adaptation according to the adaptation rules specified in the adaptation model. We analyze the dependencies between the authoring process and the functionality of the adaptation engine. From this we conclude how the authoring process can be simplified by a more powerful AE. In particular, a well-designed AE should be general purpose (i.e., not application domain specific) and should guarantee that the interpretation of the rules is deterministic, always terminates and produces the results desired by the author.
Keywords: adaptive hypermedia; user modeling; adaptive presentation; adaptive navigation; hypermedia reference model; adaptation rules

Short Papers

An Agent-Based Approach to Adaptive Hypermedia Using a Link Service BIBAFull-Text 260-263
  Christopher Bailey; Wendy Hall
This paper describes an approach to adaptive hypermedia by incorporating linkbases into an agent-based system (PAADS). The agents are built on top of an agent framework developed at Southampton University. Personal agents keep a local user model and provide adaptive navigation support. This is accomplished by extracting keywords found in the user model and through the user's browsing history, and by then replacing occurrences of those words with URL's supplied by a linkbase agent. A third agent provides the ability to query these user models through a web browser.
Adaptive Testing by Test++ BIBAFull-Text 264-267
  Maria Barra; Giuseppina Palmieri; Simona Napolitano; Vittorio Scarano; Luca Zitarosa
We present the adaptive features of Test++, an adaptive system for training and teaching on the Internet. The system integrates an adaptive training environment for personalized training and a cooperative environment for exams both accessible via Internet and a standard Java-enabled browser.
What Does the User Want to Know About Web Resources? A User Model for Metadata BIBAFull-Text 268-271
  Diana Bental; Alison Cawsey; Patrick Mc Andrew; Bruce Eddy
MIRADOR aims to create tailored descriptions which assist users in selecting relevant resources on the Web. In this paper, we distinguish between the processes of searching with a query and selection among the results of the query, and we describe a user model for metadata which supports selection. We outline how we may combine the user model with aspects of tailoring to the query, so as to produce concise and useful descriptions of resources.
Web Information Retrieval for Designing Distance Hypermedia Courses BIBAFull-Text 272-275
  Angelique Boitel; Dominique Leclet
This paper presents research work in progress that aims to desgin a Web educative information retrieval system in a Distance Education (DE) context. In order to adapt it to trainers features, experimental information searches were performed with trainers involved in a DE course. These experiments allow us to express several assumptions about trainers needs. Thus, we propose a data model and a treatment model which take into account trainers features in the Web information retrieval process.
Formative Evaluation of Adaptive CALLware: A Case Study BIBAFull-Text 276-279
  Licia Calvi
The paper presents the results of a formative evaluation concerning an on-line adaptive system to learn Italian. The purpose of this evaluation was to verify the validity of the chosen design methodology and to decide on the system's future development. The design methodology that we have adopted consisted in the use of adaptive disabling of links for a Web-based courseware. This was checked against two alternative versions of the same courseware, i.e., a static version and an adaptive version with removal of links.
How Adaptivity Affects the Development of TANGOW Web-Based Courses BIBAFull-Text 280-283
  Rosa María Carro; Estrella Pulido; Pilar Rodríguez
In this paper we describe those aspects a designer has to take into account when designing Web-based adaptive courses with TANGOW. The discussion is specifically focused on the adaptivity issues, which are not covered by any standard design methodology. The first section describes the decisions that need to be made in order to create adaptive courses by using TANGOW. Work done on some approaches to facilitate the design of Web courses is presented next. The paper finishes with some conclusions and work in progress.
An Adaptive Web Content Delivery System BIBAFull-Text 284-288
  Jinlin Chen; Yudong Yang; Hongjiang Zhang
The desktop-centric design of most of the current web contents pose many difficulties for pervasive browsing. In this paper, we present our study on the problem to support pervasive browsing in the heterogeneous environment of today's Internet. A system solution -- Adaptive Web Content Delivery (AWCD), is presented to overcome the problems existed in present web infrastructure. The system designed is extensible for the further development of Internet. The two major subsystems of AWCD, client profile learning and adaptation, are described in detail. Experiment results of our system are also shown.
Knowledge Computing Method for Enhancing the Effectiveness of a WWW Distance Education System BIBAFull-Text 289-292
  Alexandra Cristea; Toshio Okamoto
The ultimate aim of our research is a free, evolutionary, Internetbased, agent-based, long-distance teaching environment for academic English. For this purpose, we are building 2 environments: student learning and teacher courseware design environment. Here we focus on the second research direction, on constructing the teacher authorware environment (courseware management), and especially on a new method of automatic, knowledge computing-based courseware indexing that uses the AI paradigm of Concept Mapping.
Interface Adaptation to Style of User-Computer Interaction BIBAFull-Text 293-296
  Maia Dimitrova; Dimcho Boyadjiev; Nikolai Butorin
The paper presents a framework for interface adaptation to style of user-computer interaction by implicit retrieval of apriori defined independent measures of graphical, verbal, procedural and dynamic cognitive characteristics. A neural network is proposed that samples and classifies patterns of style representing data into sets of paired relations among scores for further distinct support. The approach is currently being implemented into a database system for a community mental health center.
Adaptation and Generation in a Web-Based Lisp Tutor BIBAFull-Text 297-300
  Isabel Fernández-Anta; Eva Millán; José-Luis Pérez-de-la-Cruz
This paper presents an aspect of a tool which will be developed to help teachers in the task of teaching basic Lisp programming. The paper focuses onto the problem generator module.
Collaborative Maintenance in ULYSSES BIBAFull-Text 301-304
  Maria Angela Ferrario; Keith Waters; Barry Smyth
Maintaining large-scale content repositories is a challenging issue. In this paper we describe collaborative maintenance, a novel framework to intelligently support a distributed user-driven maintenance strategy here implemented in the Ulysses online entertainment planner.
An Adaptive Open Hypermedia System on the Web BIBAFull-Text 305-310
  Giovanni Fulantelli; Riccardo Rizzo; Marco Arrigo; Rossella Corrao
A prototype of an open and adaptive hypertext learning environment on the Web is presented. The nodes of the hypertext system are sorted in clusters which are ordered on a map by a self organizing neural network. This map represents the Information Domain Model on which an overlay model of the learning goal and of the user knowledge is built up. The User Model is made up of a user knowledge model and a preference model that takes into account the user's attitude to approaching the information necessary to achieve the learning goal. The Information Domain Model allows users to add new documents to the system that are ordered by the neural network in the appropriate clusters. This maintains the consistency of the User Model and of the learning goal when the number of documents grows.
Towards an Adaptive Learners' Dictionary BIBAFull-Text 311-314
  Johann Gamper; Judith Knapp
This paper presents an ongoing research project about the development of an electronic learners' dictionary for the German and the Italian language (ELDIT). Modern psycholinguistic methods for language acquisition will be applied together with technologies for hypermedia and adaptive systems in order to ensure maximum effectiveness.
Concept Filtering and Spatial Filtering in an Adaptive Information System BIBAFull-Text 315-318
  Serge Garlatti; Sébastien Iksal
Adaptive on-line information systems are able to adapt or to personalize the hyper-document overall structure and content, the navigation help and the layout according to a user's model. SWAN aims at designing an adaptive web server for on-line information systems about nautical publications. According to a task model, concept filtering and spatial filtering are used to select the relevant information space and to sort its content.
Analysing Web Search Logs to Determine Session Boundaries for User-Oriented Learning BIBAFull-Text 319-322
  Ayse Göker; Daqing He
Incremental learning approaches based on user search activities provide a means of building adaptive information retrieval systems. To develop more effective user-oriented learning techniques for the Web, we need to be able to identify a meaningful session unit from which we can learn. Without this, we run a high risk of grouping together activities that are unrelated or perhaps not from the same user. We are interested in detecting boundaries of sequences between related activities (sessions) that would group the activities for a learning purpose. Session boundaries, in Reuters transaction logs, were detected automatically. The generated boundaries were compared with human judgements. The comparison confirmed that a meaningful session threshold for establishing these session boundaries was confined to a 11-15 minute range.
Learning User Profiles in NAUTILUS BIBAFull-Text 323-326
  Marco Gori; Marco Maggini; Enrico Martinelli; Franco Scarselli
NAUTILUS is a Web recommender system that exploits a new approach to learn user profiles. The novelty consists of using a structured representation of HTML documents that allows us to split the page into logical contexts (lists, headers, paragraphs,...). The learning algorithm is based on a new neural computational model particularly suited to process structured objects.
Lexical Chaining for Web-Based Retrieval of Breaking News BIBAFull-Text 327-330
  Paula Hatch; Nicola Stokes; Joe Carthy
This paper discusses a system for online new event detection in the domain of news articles on the web. This area is related to the Topic Detection and Tracking initiative. We evaluate two benchmark systems: The first like most current web retrieval systems, relies on term repetition to calculate document relatedness. The second attempts to perform conceptual indexing through the use of the WordNet thesaurus software. We propose a novel approach for the identification of breaking news stories, which uses a technique called lexical chaining. We believe that this technique will improve the overall performance of our web retrieval system by allowing us to encapsulate the context surrounding a word and disambiguate its senses.
Designing for Social Navigation of Food Recipes BIBAFull-Text 331-334
  Kristina Hook; Jarmo Laaksolahti; Martin Svensson; Annika Waern
Social navigation has been proposed as a means to aid users to find their way through information spaces. We present an on-line grocery store that implements several different aspects of social navigation. In an initial study, we found that social trails seem to appeal to one group of users while they alienate another group of users. We discuss the implications for design of social navigation.
A Study Comparing the Use of Shaded Text and Adaptive Navigational Support in Adaptive Hypermedia BIBAFull-Text 335-342
  Jatinder Hothi; Wendy Hall; Tim Sly
The experiment discussed in this paper was designed to investigate the usability of combining two adaptive hypermedia (AH) techniques; shading fragments of text and adaptive navigation support (ANS) by means of link annotation, within the educational paradigm. The empirical study required the implementation of three prototype applications, Application A; supported ANS and shading, Application B; supported ANS and Application C; was a non-adaptive version. An in-depth evaluation showed that ANS alone was not as useful as combining it with adaptive presentation i.e. the shading method of AH. It also showed that shading reduced the effects of cognitive overload by reducing the number of highlighted links on documents.
Layered Evaluation of Adaptive Applications and Services BIBAFull-Text 343-346
  Charalampos Karagiannidis; Demetrios G. Sampson
In this paper we address the evaluation of adaptive applications and services. We propose a layered evaluation approach, where the success of adaptation the major factor affecting the acceptability of adaptive applications-is addressed at two separate layers: (i) interaction assessment and (ii) adaptation decision making. We argue that the proposed framework constitutes an effective means for the evaluation of adaptive applications and services, providing useful information for their improvement, and facilitating generalisation of evaluation results and re-use of successful design practices.
Exploratory Activity Support Based on a Semantic Feature Map BIBAFull-Text 347-350
  Mizue Kayama; Toshio Okamoto; Alexandra I. Cristea
In this paper, we propose a framework based on a sub-symbolic approach for the support of exploratory activities in a hyper-space. By using it, it is possible to express the semantic features of the whole hyperspace and the states of exploratory activities in topological order. This approach is applied to generate the navigation information for the exploratory activity. The space explored is changed automatically by using the semantic similarities of the nodes which constitute that space. An extended self-organizing feature map is used as the semantic feature map of the hyperspace. This map is applied to express the user model and generate the navigation strategy for the user. The exploratory history of the user is mapped on it. Then, the semantic relations between nodes are shown on the map. The result reflects the exploratory state of the user, interpreted with the help of a user model.
Adaptivity in AHMED BIBAFull-Text 351-354
  Jaakko Kurhila; Erkki Sutinen
A learning environment based on adaptive hypermedia supports learners with deficits in self-instructability to navigate through a learning material. The schema used in the learning environment allows extending intelligent tutoring to the area of open learning where the learner-together with the learning environment and its designer-shares the responsibility of the learning process and actively sets its goals.
An Adaptive Document Generation Based on Matrix of Contents BIBAFull-Text 355-358
  Mona Laroussi; Pr Mohamed Ben Ahmed; Mauro Marinilli
We describe the mechanism used in the CAMELEON system to produce contents adapted to the user. The mechanism is based on matrix products using representations of users and contents by means of boolean vectors.
Logical Dimensions for the Information Provided by a Virtual Guide BIBAFull-Text 359-362
  Luisa Marucci; Fabio Paternò
If we consider most current applications we can notice a lack of support able to adapt to the different information needs that different users may have regarding a certain topic. In this paper, we present a solution that integrates a virtual assistant, able to implement adaptive support, in an adaptable application and discuss the logical dimensions of the information that can be provided by the virtual guide.
Automated Collaborative Filtering Applications for Online Recruitment Services BIBAFull-Text 363-368
  Rachael Rafter; Keith Bradley; Barry Smyth
Online recruitment services suffer from shortcomings due to traditional search techniques. Most users fail to construct queries that provide an adequate and accurate description of their (job) requirements, leading to imprecise search results. We investigate one potential solution that combines implicit profiling methods and automated collaborative filtering (ACF) techniques to build personalised query-less job recommendations. Two ACF strategies are implemented and evaluated in the JobFinder domain.
ConTexts: Adaptable Hypermedia BIBAFull-Text 369-374
  m. c. schraefel
ConTexts is an implementation of and proposed design model for an adaptable hypermedia system. ConTexts provides an easy yet powerful way to author the interactive, adaptable effects of adaptive hypermedia without Natural Language Generation modeling. We present an overview of the architecture and design methodology for ConTexts, its relationship to other technologies, and future research directions for this adaptable hypermedia.
Coherence in Modularly Composed Adaptive Learning Documents BIBAFull-Text 375-379
  Cornelia Seeberg; Achim Steinacker; Ralf Steinmetz
In this paper we suggest the Multibook approach how the gap between adaptivity and readability can be diminished. We show how a knowledge base has to be described by metadata (LOM), rhetorical-didactic relations and an underlying ontology to make it possible for the learners to build coherence from the modularly composed document.
ACE-Adaptive Courseware Environment BIBAFull-Text 380-383
  Marcus Specht
The Adaptive Courseware Environment (ACE) [1] is a WWW-based tutoring framework which combines methods of knowledge representation, instructional planning, and adaptive media generation to deliver individualized courseware over the WWW. In this paper we like to introduce new cooperative components integrated in the ACE framework that support cooperative knowledge/data collection, expert finding and learner group organization. All these new cooperative components are realized with modified components of the BSCW environment [2]. A pedagogical agent in ACE is able to make recommendations for switching between individual and cooperative learning activities and find advanced or expert students in relation to one's own profile.
The Adaptive University Calendar BIBAFull-Text 384-387
  Manfred Stede; Stephan Koch
Each term, university students consult the new calendar for deciding which courses to take in order to advance their studies. At present, university calendars are largely static documents that do not make use of adaptive hypertext technology. We are currently developing a prototypical system that offers course information tailored to the specific student, on the basis of a user profile and the navigation history of the current system session. Documents are thus partly generated dynamically, and hyperlinks into the document space created appropriately. We describe the idea, the user scenario with its roles for adaptivity, and the technical realization, which is currently under way.
Sense-Based User Modelling for Web Sites BIBAFull-Text 388-391
  Carlo Strapparava; Bernardo Magnini; Anna Stefani
SiteIF is a personal agent for a news web site that learns user's interests from the requested pages that are analyzed to generate or to update a model of the user. Exploiting this model, the system anticipates which documents in the web site could be interesting for the user. Using MultiWordNet, a multilingual extension of WordNet, a content-based user model is built as a semantic network whose nodes, independent from the language, represent the word sense frequency rather then word frequency.
Generating Personal Travel Guides from Discourse Plans BIBAFull-Text 392-395
  Ross Wilkinson; Shijian Lu; François Paradis; Cécile Paris; Stephen Wan; MingFang Wu
This paper describes a system that delivers travel guides tailored to individual needs. It does so by integrating a discourse planner with a system for querying the web and generating synthesised web pages using document prescriptions. We show by way of example how a user model can lead to a personal travel guide, and show what this might look like in different media. We briefly describe the investigations we are undertaking to determine the utility of such approaches.

Doctoral Consortium Papers

Distributed Systems for Group Adaptivity on the Web BIBAFull-Text 396-402
  Maria Barra
This paper describes the project of an Adaptive System for Group navigation on the Web. The objective of the project is to provide collaborative and adaptive navigation to users groups sharing a "common interest" on the Web. The goal of the collaborative navigation is to promote the knowledge and information exchange among users having common objectives.
Open Multimedia Environment to Retrieve and Organise Documents: An Adaptive Web-Based IR System in the Field of Textile and Clothing Industry BIBAFull-Text 403-408
  Cristiano Chesi; Francesca Rizzo
Computer based information repositories are becoming larger and more diverse. In this context the need for an effective information retrieval system is related not only to the efficiency of the retrieval process but also to its compatibility to support information seekers during a typical problem solving activity (Marchionini 1995). OMERO project (national research project, n. 41902) perspective considers information seeking as a problem solving activity (Rumiati 1990) that depends on communication acts.
Researching Adaptive Instruction BIBAFull-Text 409-414
  Juan E. Gilbertau; Chia Y. Han
Human tutors have the ability to explain concepts several different ways using multiple instruction methods. Typically, instruction methods are used when students are having difficulty learning a concept. Intelligent tutoring systems are expert systems [5] where the expert is a tutor. Ideally, an intelligent tutoring system should have the ability to explain concepts several different ways using a knowledge base of multiple instruction methods [3,4,6,7]. The technique of providing instruction using multiple instruction methods during a single tutoring session is "Adaptive Instruction". An intelligent tutoring system that provides adaptive instruction has been developed, Arthur[2]. Arthur has been used in a research experiment with human subjects. The results of this experiment are discussed in the section below.
A Modular Approach for User Modelling BIBAFull-Text 415-420
  Ilaria Torre
Adaptive hypermedia systems are spreading widely in these last years, but each application uses its own models and techniques. What I am studying is the possibility of developing a framework for user modelling in adaptive systems and in particular of creating a library of stereotypes for representing the different dimensions of the user models.