| Assessment of User Affective and Belief States for Interface Adaptation: Application to an Air Force Pilot Task | | BIBAK | Full-Text | 1-47 | |
| Eva Hudlicka; Michael D. McNeese | |||
| We describe an Affect and Belief Adaptive Interface System (ABAIS) designed
to compensate for performance biases caused by users' affective states and
active beliefs. The ABAIS architecture implements an adaptive methodology
consisting of four steps: sensing/inferring user affective state and
performance-relevant beliefs; identifying their potential impact on
performance; selecting a compensatory strategy; and implementing this strategy
in terms of specific GUI adaptations. ABAIS provides a generic adaptive
framework for integrating a variety of user assessment methods (e.g.
knowledge-based, self-reports, diagnostic tasks, physiological sensing), and
GUI adaptation strategies (e.g. content- and format-based). The ABAIS
performance bias prediction is based on empirical findings from emotion
research combined with detailed knowledge of the task context. The initial
ABAIS prototype was demonstrated in the context of an Air Force combat task,
used a knowledge-based approach to assess the pilot's anxiety level, and
adapted to the pilot's anxiety and belief states by modifying selected cockpit
instrument displays in response to detected changes in these states. Keywords: adaptive interface; affect adaptation; affect assessment; affective
computing; aviation; human-computer interaction; user modeling | |||
| Modeling Multimodal Expression of User's Affective Subjective Experience | | BIBAK | Full-Text | 49-84 | |
| Nadia Bianchi-Berthouze; Christine L. Lisetti | |||
| With the growing importance of information technology in our everyday life,
new types of applications are appearing that require the understanding of
information in a broad sense. Information that includes affective and
subjective content plays a major role not only in an individual's cognitive
processes but also in an individual's interaction with others. We identify
three key points to be considered when developing systems that capture
affective information: embodiment (experiencing physical reality), dynamics
(mapping experience and emotional state with its label) and adaptive
interaction (conveying emotive response, responding to a recognized emotional
state). We present two computational systems that implement those principles:
MOUE (Model Of User Emotions) is an emotion recognition system that recognizes
the user's emotion from his/her facial expressions, and from it, adaptively
builds semantic definitions of emotion concepts using the user's feedback; MIKE
(Multimedia Interactive Environment for Kansei communication) is an interactive
adaptive system that, along with the user, co-evolves a language for
communicating over subjective impressions. Keywords: affect; embodiment; emotion; interaction; perception; subjective experience | |||
| Rosalind Picard: Affective Computing | | BIB | Full-Text | 85-89 | |
| Annika Waern | |||
| Ana Paiva (ed.): Affective Interactions: Towards a New Generation of Computer Interfaces | | BIB | Full-Text | 91-96 | |
| Kristina Höök | |||
| Introduction to the Special Issue on Empirical Evaluation of User Models and User Modeling Systems | | BIB | Full-Text | 105-109 | |
| David N. Chin; Martha E. Crosby | |||
| Designing and Evaluating an Adaptive Spoken Dialogue System | | BIBAK | Full-Text | 111-137 | |
| Diane J. Litman; Shimei Pan | |||
| Spoken dialogue system performance can vary widely for different users, as
well for the same user during different dialogues. This paper presents the
design and evaluation of an adaptive version of TOOT, a spoken dialogue system
for retrieving online train schedules. Based on rules learned from a set of
training dialogues, adaptive TOOT constructs a user model representing whether
the user is having speech recognition problems as a particular dialogue
progresses. Adaptive TOOT then automatically adapts its dialogue strategies
based on this dynamically changing user model. An empirical evaluation of the
system demonstrates the utility of the approach. Keywords: adaptive spoken dialogue systems; hypothesis testing for the effectiveness
of adaptations; PARADISE for evaluating performance measures; speech
recognition; user model acquisition via machine learning | |||
| User Models and User Physical Capability | | BIBAK | Full-Text | 139-169 | |
| Simeon Keates; Patrick Langdon | |||
| Current interface design practices are based on user models and descriptions
derived almost exclusively from studies of able-bodied users (Keates et al.,
1999). However, such users are only one point on a wide and varied scale of
physical capabilities.
Users with a number of different physical impairment conditions have the same desire to use computers as able-bodied people (Busby, 1997), but cannot cope with most current computer access systems (Edwards, 1995). It is important to identify the differences in interaction for users of differing physical capability, because the border between the labels 'able-bodied' and 'motion-impaired' users is becoming increasingly blurred as the generation of computer users inexorably becomes older and physically less capable. If user models are to retain their relevance, then they have to be able to reflect users' physical capabilities (Stary, 1997). Through empirical studies, this paper will show that there are very important differences between those with motion-impairments, whether elderly or disabled, and able-bodied users when they interact with computers. It attempts to quantify where those differences occur in the interaction cycle with the use of a very straightforward user model, the Model Human Processor (MHP) (Card, Moran and Newell, 1983), which describes interaction purely in terms of perception, cognition and motor component times. Although this model is simplistic compared to the more recent sophisticated models, it affords a simple and valuable insight into interaction cycles and offers a building block on which to base more comprehensive models. This work is predicated on the idea that the use of this model in detailed analysis of the basic interaction cycle will provide a means for studying motion impairment at both an individual and general level. Keywords: model human processor; motion-impaired users | |||
| Evaluating Comprehension-Based User Models: Predicting Individual User Planning and Action | | BIBAK | Full-Text | 171-205 | |
| Young Woo Sohn; Stephanie M. Doane | |||
| Described is a program of research that uses rigorous methods to evaluate
models of user cognition and action based on the construction-integration
architecture of comprehension (Doane and Sohn, 2000; Kintsch, 1988; 1998). The
models interrelate user environmental information, background knowledge, and
current goals, and then spread activation throughout the interrelated
information to simulate UNIX user command productions, aviation pilot eye
fixations and control movements during flight, and army personnel intelligence
planning. Models of individuals in the complex interactive environments are
tested for descriptive as well as predictive validity. Comparisons of model and
human empirical data have resulted in a high degree of agreement, validating
the ability of the comprehension-based architecture to support models that can
predict user performance. Evaluation methods are detailed and the importance of
evaluative rigor is discussed. Keywords: cognitive models; evaluations of models; goodness of fit; predictive
validity | |||
| The Use of a Co-operative Student Model of Learner Characteristics to Configure a Multimedia Application | | BIBAK | Full-Text | 207-241 | |
| Trevor Barker; Sara Jones; Carol Britton | |||
| This paper describes an investigation into the ways in which learning using
a multimedia application can be supported and enhanced by means of a simple
co-operative student model of learner characteristics. This paper reports the
design, implementation and evaluation of an individually configurable
multimedia learning application, based upon such a model.
A multimedia learning application was developed that presented information differentially based upon the individual characteristics of learners, held in the student model. The characteristics employed in the model were language level, cognitive style, task and question levels, and help level. Small groups of learners followed the multimedia course in learning centres located in colleges in the UK. A Grounded Theory study was carried out in order to understand the many and complex interactions that took place between learners, tutors and the learning environment. Stages in the Grounded Theory method are described and some qualitative data is presented. It was possible to conclude from these, that the quality of learning for individuals was improved by the use of the co-operative student model. Quantitative data is presented to support this view and where possible, to relate performance on the multimedia learning application to the student model configuration. Keywords: CAL; evaluation; global descriptors; Grounded Theory; multimedia; Student
model | |||
| Using Evaluation to Shape ITS Design: Results and Experiences with SQL-Tutor | | BIBAK | Full-Text | 243-279 | |
| Antonija Mitrovic; Brent Martin; Michael Mayo | |||
| This paper presents the results of three evaluation studies performed during
1998 and 1999 on SQL-Tutor, an intelligent tutoring system for the SQL database
language. We have evaluated the system in the context of genuine courses, and
used the results to further refine the system. The main goal of our research
has been the exploration and extension of Constraint-Based Modeling (CBM), a
student modeling approach proposed by Ohlsson (1994). SQL-Tutor provided us
with experiences of using CBM, and we used it to extend the approach in several
important ways. The main goal of all three evaluation studies was to determine
how well CBM supported student learning. We have obtained positive results. The
students who learnt with SQL-Tutor in the first study performed significantly
better than those who did not when assessed by a subsequent classroom
examination. Furthermore, the analysis of students' learning shows that CBM has
a sound psychological foundation.
Besides the evaluation of CBM, we also evaluated the improvements in terms of student assessments of the usefulness of the system and evaluated various techniques used in SQL-Tutor. In the second study, we evaluated the effectiveness of feedback provided to the students. This study showed that high-level advice is most beneficial to students' learning. The focus of the third study was different. We extended CBM to support long-term modeling of student knowledge, and used this extension to develop an adaptive problem-selection strategy. The study revealed the benefits of this strategy in comparison with a simple heuristic strategy. We also reflect on our experiences in evaluating SQL-Tutor. Keywords: constraint based modeling; evaluation; intelligent tutoring systems;
pedagogical decision making; probabilistic student model; student modeling | |||
| A Bayesian Diagnostic Algorithm for Student Modeling and its Evaluation | | BIBAK | Full-Text | 281-330 | |
| Eva Millán; José Luis Pérez-de-la-Cruz | |||
| In this paper, we present a new approach to diagnosis in student modeling
based on the use of Bayesian Networks and Computer Adaptive Tests. A new
integrated Bayesian student model is defined and then combined with an Adaptive
Testing algorithm. The structural model defined has the advantage that it
measures students' abilities at different levels of granularity, allows
substantial simplifications when specifying the parameters (conditional
probabilities) needed to construct the Bayesian Network that describes the
student model, and supports the Adaptive Diagnosis algorithm. The validity of
the approach has been tested intensively by using simulated students. The
results obtained show that the Bayesian student model has excellent performance
in terms of accuracy, and that the introduction of adaptive question selection
methods improves its behavior both in terms of accuracy and efficiency. Keywords: adaptive testing; Bayesian networks; student modeling | |||
| Hybrid Recommender Systems: Survey and Experiments | | BIBAK | Full-Text | 331-370 | |
| Robin Burke | |||
| Recommender systems represent user preferences for the purpose of suggesting
items to purchase or examine. They have become fundamental applications in
electronic commerce and information access, providing suggestions that
effectively prune large information spaces so that users are directed toward
those items that best meet their needs and preferences. A variety of techniques
have been proposed for performing recommendation, including content-based,
collaborative, knowledge-based and other techniques. To improve performance,
these methods have sometimes been combined in hybrid recommenders. This paper
surveys the landscape of actual and possible hybrid recommenders, and
introduces a novel hybrid, EntreeC, a system that combines knowledge-based
recommendation and collaborative filtering to recommend restaurants. Further,
we show that semantic ratings obtained from the knowledge-based part of the
system enhance the effectiveness of collaborative filtering. Keywords: case-based reasoning; collaborative filtering; recommender systems | |||
| Using Bayesian Networks to Manage Uncertainty in Student Modeling | | BIBAK | Full-Text | 371-417 | |
| Cristina Conati; Abigail Gertner | |||
| When a tutoring system aims to provide students with interactive help, it
needs to know what knowledge the student has and what goals the student is
currently trying to achieve. That is, it must do both assessment and plan
recognition. These modeling tasks involve a high level of uncertainty when
students are allowed to follow various lines of reasoning and are not required
to show all their reasoning explicitly. We use Bayesian networks as a
comprehensive, sound formalism to handle this uncertainty. Using Bayesian
networks, we have devised the probabilistic student models for Andes, a
tutoring system for Newtonian physics whose philosophy is to maximize student
initiative and freedom during the pedagogical interaction. Andes' models
provide long-term knowledge assessment, plan recognition, and prediction of
students' actions during problem solving, as well as assessment of students'
knowledge and understanding as students read and explain worked out examples.
In this paper, we describe the basic mechanisms that allow Andes' student
models to soundly perform assessment and plan recognition, as well as the
Bayesian network solutions to issues that arose in scaling up the model to a
full-scale, field evaluated application. We also summarize the results of
several evaluations of Andes which provide evidence on the accuracy of its
student models. Keywords: student modelling; Intelligent Tutoring Systems; dynamic Bayesian networks | |||
| Book Review: User Interfaces for All: Concepts, Methods and Tools | | BIB | Full-Text | 419-420 | |
| Bonnie A. Nardi | |||