Showing papers in "User Modeling and User-adapted Interaction in 2005"

User Modeling and User-Adapted Interaction

Abstract: Please provide a short abstract of 100 to 250 words. The abstract should not contain any undefined abbreviations or unspecified references. Work planned but not completed should not appear in the abstract.

Exploiting Query Repetition and Regularity in an Adaptive Community-Based Web Search Engine

Abstract: Search engines continue to struggle with the challenges presented by Web search: vague queries, impatient users and an enormous and rapidly expanding collection of unmoderated, heterogeneous documents all make for an extremely hostile search environment. In this paper we argue that conventional approaches to Web search -- those that adopt a traditional, document-centric, information retrieval perspective -- are limited by their refusal to consider the past search behaviour of users during future search sessions. In particular, we argue that in many circumstances the search behaviour of users is repetitive and regular; the same sort of queries tend to recur and the same type of results are often selected. We describe how this observation can lead to a novel approach to a more adaptive form of search, one that leverages past search behaviours as a means to re-rank future search results in a way that recognises the implicit preferences of communities of searchers. We describe and evaluate the I-SPY search engine, which implements this approach to collaborative, community-based search. We show that it offers potential improvements in search performance, especially in certain situations where communities of searchers share similar information needs and use similar queries to express these needs. We also show that I-SPY benefits from important advantages when it comes to user privacy. In short, we argue that I-SPY strikes a useful balance between search personalization and user privacy, by offering a unique form of anonymous personalization, and in doing so may very well provide privacy-conscious Web users with an acceptable approach to personalized search.

Personalization and Context Management

Abstract: Supporting the individual user in his working, learning, or information access is one of the main goals of user modeling. Personal or group user models make it possible to represent and use information about preferences, knowledge, abilities, emotional states, and many other characteristics of a user to adapt the user experience and support. Nowadays, the disappearing computer enables the user to access her information from a variety of personal and public displays and devices. To support a new generation of contextualized and personalized information and services, this paper addresses the problem of context management. Context management is a new approach to the design of context-aware systems in ubiquitous computing that combines personalization and contextualization. The presented framework for context management integrates user modeling and context modeling, which can benefit from each other and give rise to more valid models for personalized and contextualized information delivery. The paper will introduce a base framework and tools for designing context-aware applications and decompose the underlying framework into its foundational components. As two illustrative application cases, the paper discusses implementations of an intelligent advertisement board and an audio-augmented museum environment.

Knowledge Discovery with Genetic Programming for Providing Feedback to Courseware Authors

Abstract: We introduce a methodology to improve Adaptive Systems for Web-Based Education. This methodology uses evolutionary algorithms as a data mining method for discovering interesting relationships in students' usage data. Such knowledge may be very useful for teachers and course authors to select the most appropriate modifications to improve the effectiveness of the course. We use Grammar-Based Genetic Programming (GBGP) with multi-objective optimization techniques to discover prediction rules. We present a specific data mining tool that can help non-experts in data mining carry out the complete rule discovery process, and demonstrate its utility by applying it to an adaptive Linux course that we developed.

User-Centred Design of Flexible Hypermedia for a Mobile Guide: Reflections on the HyperAudio Experience

Abstract: A user-centred design approach involves end-users from the very beginning. Considering users at the early stages compels designers to think in terms of utility and usability and helps develop a system based on what is actually needed. This paper discusses the case of HyperAudio, a context-sensitive adaptive and mobile museum guide developed in the late 1990s. User requirements were collected via a survey to understand visitors' profiles and visit styles in natural science museums. The knowledge acquired supported the specification of system requirements, helping define the user model, data structure and adaptive behaviour of the system. User requirements guided the design decisions on what could be implemented by using simple adaptable triggers, and what instead needed more sophisticated adaptive techniques. This is a fundamental choice when all the computation must be done on a PDA. Graphical and interactive environments for developing and testing complex adaptive systems are discussed as a further step in an iterative design process that considers the user interaction to be the central point. This paper discusses how such an environment allows designers and developers to experiment with different system behaviours and to widely test it under realistic conditions by simulating the actual context evolving over time. The understanding gained in HyperAudio is then considered from the perspective of later developments: our findings still appers to be valid despite the time that had passed.

Ontology-Based User Modeling in an Augmented Audio Reality System for Museums

Abstract: Ubiquitous computing is a challenging area that allows us to further our understanding and techniques of context-aware and adaptive systems. Among the challenges is the general problem of capturing the larger context in interaction from the perspective of user modeling and human---computer interaction (HCI). The imperative to address this issue is great considering the emergence of ubiquitous and mobile computing environments. This paper provides an account of our addressing the specific problem of supporting functionality as well as the experience design issues related to museum visits through user modeling in combination with an audio augmented reality and tangible user interface system. This paper details our deployment and evaluation of ec(h)o --- an augmented audio reality system for museums. We explore the possibility of supporting a context-aware adaptive system by linking environment, interaction objects and users at an abstract semantic level instead of at the content level. From the user modeling perspective ec(h)o is a knowledge-based recommender system. In this paper we present our findings from user testing and how our approach works well with an audio and tangible user interface within a ubiquitous computing system. We conclude by showing where further research is needed.

Exploring Issues of User Model Transparency and Proactive Behaviour in an Office Environment Control System

Abstract: It is important that systems that exhibit proactive behaviour do so in a way that does not surprise or frustrate the user. Consequently, it is desirable for such systems to be both personalised and designed in such a way as to enable the user to scrutinise her user model (part of which should hold the rules describing the behaviour of the system). This article describes on-going work to investigate the design of a prototype system that can learn a given user's behaviour in an office environment in order to use the inferred rules to populate a user model and support appropriate proactive behaviour (e.g. turning on the user's fan under appropriate conditions). We explore the tension between user control and proactive services and consider issues related to the design of appropriate transparency with a view to supporting user comprehensibility of system behaviour. To this end, our system enables the user to scrutinise and possibly over-ride the `IF-THEN' rules held in her user model. The system infers these rules from the context history (effectively a data set generated using a variety of sensors) associated with the user by using a fuzzy-decision-tree-based algorithm that can provide a confidence level for each rule in the user model. The evolution of the system has been guided by feedback from a number of real-life users in a university department. A questionnaire study has yielded supplementary results concerning the extent to which the approach taken meets users' expectations and requirements.

Consistent Modelling of Users, Devices and Sensors in a Ubiquitous Computing Environment

Abstract: This paper describes the use of an accretion-resolution user modelling representation to model people, places and objects We explain the motivation for the key properties of the representation, especially those of particular importance for ubiquitous computing: firstly, for flexibility in interpreting the typically noisy and potentially conflicting evidence about users' locations; secondly, to support users in scrutinising their user model, the processes that determine its contents and the way that it is used in the ubiquitous computing environment A novel and important aspect of this work is our extension of the representation beyond modelling just users, using it also to represent the other elements such as devices, sensors, rooms and buildings We illustrate our approach in terms of models we have been building for a system which enables users to gain personalised information about the sensors and services in a ubiquitous computing environment We report experiments on the scalability and the management of inconsistency in modelling of location, based on accretion-resolution

Using Dialogue Features to Predict Trouble During Collaborative Learning

Abstract: A web-based, collaborative distance-learning system that will allow groups of students to interact with each other remotely and with an intelligent electronic agent that will aid them in their learning has the potential for improving on-line learning. The agent would follow the discussion and interact with the participants when it detects learning trouble of some sort, such as confusion about the problem they are working on or a participant who is dominating the discussion or not interacting with the other participants. In order to recognize problems in the dialogue, we investigated conversational elements that can be utilized as predictors for effective and ineffective interaction between human students. These elements can serve as the basis for student and group models. In this paper, we discuss group interaction during collaborative learning, our representation of participant dialogue, and the statistical models we are using to determine the role being played by a participant at any point in the dialogue and the effectiveness of the group. We also describe student and group models that can be built using conversational elements and discuss one set that we built to illustrate their potential value in collaborative learning.

User Modeling in Spoken Dialogue Systems to Generate Flexible Guidance

Abstract: We address the issue of appropriate user modeling to generate cooperative responses to users in spoken dialogue systems. Unlike previous studies that have focused on a user's knowledge, we propose more generalized modeling. We specifically set up three dimensions for user models: the skill level in use of the system, the knowledge level about the target domain, and the degree of urgency. Moreover, the models are automatically derived by decision tree learning using actual dialogue data collected by the system. We obtained reasonable accuracy in classification for all dimensions. Dialogue strategies based on user modeling were implemented on the Kyoto City Bus Information System that was developed at our laboratory. Experimental evaluations revealed that the cooperative responses adapted to each subject type served as good guides for novices without increasing the duration dialogue lasted for skilled users.

Modeling Elementary Cognitive Abilities for Adaptive Hypermedia Presentation

Abstract: The adaptation of hypermedia can be carried out at three levels, namely the content, navigation and presentation level. The presentation level is the least studied of the three, apparently because it refers to user properties that are not easy to model. In this paper, we present a new approach to modeling cognitive abilities that relies on basic mental functionalities. We describe the Cognitive User Modeling for Adaptive Presentation of Hyper-Documents (CUMAPH) environment, which mainly provides an authoring tool and an adaptation engine. The aim of this environment is to adapt a hyper-document presentation by selecting the elements that best fit the user cognitive profile. Its architecture is based on four main components: a cognitive user model, a hyper-document builder, an adaptation engine and a generic style sheet. To validate our approach, we designed an innovative protocol and conducted an experimental study involving 39 students. The first results show that an adaptive presentation can significantly increase the efficiency of hypermedia presentations.

Personal Content Recommender Based on a Hierarchical User Model for the Selection of TV Programmes

Abstract: In this paper we present our approach to user modeling for a personalized selection of multimedia content tested on a corpus of TV programmes. The idea of this approach is to classify content (TV programmes) based on the calculation of similarities between the description of content and the user model for each description attribute. Calculated similarities are then combined into a classification decision using the Support Vector Machines. The basis for the calculation of similarities is a hierarchical structure of the user model, overlaid upon a taxonomy of TV programme genres. Preliminary results show that it works well with a varying quality of content descriptions including incomplete genre classification and arbitrary number of description attributes. The evaluation of the system performance was based on content described using the TV-Anytime standard, but the approach can be adapted for search of other types of content with multi-attribute descriptions.

Tailoring Automatically Generated Hypertext

Abstract: This paper describes an approach for tailoring the content and structure of automatically generated hypertext. The implemented system HYLITE is based on applied Natural Language Generation (NLG) techniques, a re-usable user modelling component (VIEWGEN), and a flexible architecture with module feedback. The user modelling component is used by the language generation modules to adapt the hypertext content and links to user beliefs and preferences and to the previous interaction. Unlike previous adaptive NLG systems, which have their own, application-specific user models, HYLITE has re-used a generic agent modelling framework (VIEWGEN) instead. Apart from avoiding the development costs of a new model, this also enabled a more extendable system architecture. Another distinct feature of our approach is making NLG techniques adaptable by the user, i.e., providing users with control over the user model and the hypertext adaptivity.

A Probabilistic Approach for Argument Interpretation

Abstract: We describe a probabilistic approach for the interpretation of user arguments, and investigate the incorporation of different models of a user's beliefs and inferences into this mechanism. Our approach is based on the tenet that the interpretation intended by the user is that with the highest posterior probability. This approach is implemented in a computer-based detective game, where the user explores a virtual scenario, and constructs an argument for a suspect's guilt or innocence. Our system receives as input an argument entered through a web interface, and produces an interpretation in terms of its underlying knowledge representation -- a Bayesian network. This interpretation may differ from the user's argument in its structure and in its beliefs in the argument propositions. We conducted a synthetic evaluation of the basic interpretation mechanism, and a user-based evaluation which assesses the impact of the different user models. The results of both evaluations were encouraging, with the system generally producing argument interpretations our users found acceptable.

Preface to the Special Issue on User Modeling in Ubiquitous Computing

Abstract: The field of user modeling has come up with many techniques for modeling and adapting to computer users, for example, to their preferences, goals, and intentions, as well as to their cognitive and affective states. Until relatively recently, these methods were restricted to desktop systems, in which the user’s external context could largely be neglected. With the increasing ubiquity of mobile and embedded devices, it has become clear that in many cases the recognition and modeling of the user’s external context is essential. Coming from the other direction, ubiquitous computing has generated many approaches to recognizing and modeling a user’s context, for example his or her location, physical environment, and social environment. But in most cases there has been no explicit modeling of users themselves. Recently, an increasing number of researchers in this area have taken into account the fact that the external context alone may not determine the most appropriate adaptation to the user. So they have worked on methods for recognizing and adapting to aspects of the user such as their activities, general interests, and current information needs. We believe that the field of user modeling can contribute significantly to the enhancement of the effectiveness and usability of ubiquitous computing systems. In turn, the field of ubiquitous computing, by building the technological basis for mobile and migrating systems, is offering the user modeling community opportunities to apply their methods to novel types of systems, extending the methods themselves in the process. The major conferences in both fields – the series User Modeling, Intelligent User Interfaces, Ubicomp and Pervasive, have seen several invited talks, technical sessions, and workshops during recent years that can be seen as relevant to both user modeling and ubiquitous computing. The time seems to be ripe to bring these two communities even closer together – hence this special issue. Although each article in the special issue represents a unique perspective, the articles can be divided roughly into two groups:

Combining experimental observations and modelling in investigating feedback and emotions in repeated selection tasks

Abstract: People seem to learn tasks even without formal training. This can be modelled as the outcome of a feedback system that accumulates experience. In this paper we investigate such a feedback system, following an iterative research approach. A feedback loop is specified that is detailed using contemporary ideas on human behaviour. The resulting model is investigated in an empirical study. Finally, we consider a computational mechanism to explain the results. This approach is aimed at understanding how a feedback mechanism might work rather than at observing its outcomes. In this paper, we study the approach through adjustments in card selections in a game consisting of repeated card choices. Playing this game, participants do not know what rules determine gains and losses. Therefore there is some tension between exploring the options and achieving immediate profit. To decide in such situations it is argued that often evaluations below the level of conscious awareness, such as affect, play an important role. The results support the hypothesis that participants would draw better cards as the game progressed. There is some evidence that emotions are involved, since the hypothesis that profit and emotions are correlated is confirmed. Further evidence that formal logic is not sufficient follows from the observed effects of music on card selections. In the second part of the paper the aim is to understand the results from a computational point of view. Four possible ways of integrating feedback into a decision criterion are compared. Using one of these mechanisms, a computational model is investigated that might describe the role of music in card selection. Although there are limitations to both the empirical and computational findings, the chosen approach indicates that computational modelling of experiential appraisal, at a preconscious level, and the effect of external factors, such as music, is in principle feasible, and can lead to a research agenda aimed at understanding such phenomena.

A Model to Disambiguate Natural Language Parses on the Basis of User Language Proficiency: Design and Evaluation

Abstract: This paper discusses the design and evaluation of an implemented user model in ICICLE, an instruction system for users writing in a second language. We show that in the task of disambiguating natural language parses, a blended model combining overlay techniques with user stereotyping representing typical linguistic acquisition sequences captures user individuality while supplementing incomplete information with stereotypic reasoning

Preface to the Special Issue on Language-Based Interaction

Abstract: Natural Language (NL) has been linked with User Modeling since the inception of the User Modeling field. In fact, the first User Modeling workshop, which was held in Maria Laach, Germany, in 1986, focused on the application of User Mod-eling to dialogue systems (Kobsa and Wahlster, 1989). These applications were primarily motivated by the insight that understanding people's utterances involves building and consulting a model of the interlocutor, and generating felicitous discourse requires taking into account a model of the addressee. Thus, the applications of the time focused on three main language-related aspects of user modeling: (1) building user models from NL utterances, e. (3) consulting user models to generate discourse that takes into account some aspect of These applications were typically knowledge intensive – they relied on plan libraries and applied rule-based reasoning to infer user models or generate discourse. Since then, the applications in the field of User Modeling have expanded significantly. User Modeling now includes new applications, such as recommender systems, e-commerce, mobile and ubiquitous systems, personalized TV, and group modeling. This expansion has coincided with a methodological shift in the Artificial Intelligence community towards techniques which have a strong proba-bilistic or empirical basis, such as Bayesian networks and machine learning techniques. This shift in turn has been accompanied by a requirement for a rigorous evaluation of developed systems. The articles in this Special Issue reflect this trend with respect to systems that address the three main aspects at the intersection of User Modeling and NL. Goodman et al. use machine learning techniques in a system that builds models of user interactions, and Komatani et al. apply these techniques to build models of users of a spoken dialogue system. Michaud et al. also build user models, but with a rule-based approach, while Bontcheva and Wilks apply rules for NL generation. Finally, Zukerman and George employ probabilistic methods in a discourse interpretation system that consults a user model.