| Introduction to special Issue on 'Affective modeling and adaptation' | | BIBA | Full-Text | 1-9 | |
| Sandra Carberry; Fiorella de Rosis | |||
| We are all ruled in what we do by impulses; and these impulses are so organized that our actions in general serve for our self preservation and that of the race. Hunger, love, pain, fear are some of those inner forces which rule the individual's instinct for self preservation. At the same time, as social beings, we are moved in the relations with our fellow beings by such feelings as sympathy, pride, hate, need for power, pity and so on. -- Albert Einstein, 1950 | |||
| The relative impact of student affect on performance models in a spoken dialogue tutoring system | | BIBAK | Full-Text | 11-43 | |
| Kate Forbes-Riley; Mihai Rotaru | |||
| We hypothesize that student affect is a useful predictor of spoken dialogue
system performance, relative to other parameters. We test this hypothesis in
the context of our spoken dialogue tutoring system, where student learning is
the primary performance metric. We first present our system and corpora, which
have been annotated with several student affective states, student correctness
and discourse structure. We then discuss unigram and bigram parameters derived
from these annotations. The unigram parameters represent each annotation type
individually, as well as system-generic features. The bigram parameters
represent annotation combinations, including student state sequences and
student states in the discourse structure context. We then use these parameters
to build learning models. First, we build simple models based on correlations
between each of our parameters and learning. Our results suggest that our
affect parameters are among our most useful predictors of learning,
particularly in specific discourse structure contexts. Next, we use the
PARADISE framework (multiple linear regression) to build complex learning
models containing only the most useful subset of parameters. Our approach is a
value-added one; we perform a number of model-building experiments, both with
and without including our affect parameters, and then compare the performance
of the models on the training and the test sets. Our results show that when
included as inputs, our affect parameters are selected as predictors in most
models, and many of these models show high generalizability in testing. Our
results also show that overall, the affect-included models significantly
outperform the affect-excluded models. Keywords: Emotional speech; Discourse structure; Affective user modeling; Multiple
linear regression; Adaptive spoken dialogue systems; Tutorial dialogue systems;
System performance modeling | |||
| Automatic detection of learner's affect from conversational cues | | BIBAK | Full-Text | 45-80 | |
| Sidney K. D'Mello; Scotty D. Craig | |||
| We explored the reliability of detecting a learner's affect from
conversational features extracted from interactions with AutoTutor, an
intelligent tutoring system (ITS) that helps students learn by holding a
conversation in natural language. Training data were collected in a learning
session with AutoTutor, after which the affective states of the learner were
rated by the learner, a peer, and two trained judges. Inter-rater reliability
scores indicated that the classifications of the trained judges were more
reliable than the novice judges. Seven data sets that temporally integrated the
affective judgments with the dialogue features of each learner were
constructed. The first four datasets corresponded to the judgments of the
learner, a peer, and two trained judges, while the remaining three data sets
combined judgments of two or more raters. Multiple regression analyses
confirmed the hypothesis that dialogue features could significantly predict the
affective states of boredom, confusion, flow, and frustration. Machine learning
experiments indicated that standard classifiers were moderately successful in
discriminating the affective states of boredom, confusion, flow, frustration,
and neutral, yielding a peak accuracy of 42% with neutral (chance=20%) and 54%
without neutral (chance=25%). Individual detections of boredom, confusion,
flow, and frustration, when contrasted with neutral affect, had maximum
accuracies of 69, 68, 71, and 78%, respectively (chance=50%). The classifiers
that operated on the emotion judgments of the trained judges and combined
models outperformed those based on judgments of the novices (i.e., the self and
peer). Follow-up classification analyses that assessed the degree to which
machine-generated affect labels correlated with affect judgments provided by
humans revealed that human-machine agreement was on par with novice judges
(self and peer) but quantitatively lower than trained judges. We discuss the
prospects of extending AutoTutor into an affect-sensing ITS. Keywords: Affect detection; Human-computer interaction; Human-computer dialogue;
Dialogue features; Discourse markers; Conversational cues; Intelligent Tutoring
Systems; AutoTutor | |||
| Modeling self-efficacy in intelligent tutoring systems: An inductive approach | | BIBAK | Full-Text | 81-123 | |
| Scott W. McQuiggan; Bradford W. Mott | |||
| Self-efficacy is an individual's belief about her ability to perform well in
a given situation. Because self-efficacious students are effective learners,
endowing intelligent tutoring systems with the ability to diagnose
self-efficacy could lead to improved pedagogy. Self-efficacy is influenced by
(and influences) affective state. Thus, physiological data might be used to
predict a student's level of self-efficacy. This article investigates an
inductive approach to automatically constructing models of self-efficacy that
can be used at runtime to inform pedagogical decisions. It reports on two
complementary empirical studies. In the first study, two families of
self-efficacy models were induced: a static self-efficacy model, learned solely
from pre-test (non-intrusively collected) data, and a dynamic self-efficacy
model, learned from both pre-test data as well as runtime physiological data
collected with a biofeedback apparatus. In the second empirical study, a
similar experimental design was applied to an interactive narrative-centered
learning environment. Self-efficacy models were induced from combinations of
static and dynamic information, including pre-test data, physiological data,
and observations of student behavior in the learning environment. The highest
performing induced naïve Bayes models correctly classified 85.2% of
instances in the first empirical study and 82.1% of instances in the second
empirical study. The highest performing decision tree models correctly
classified 86.9% of instances in the first study and 87.3% of instances in the
second study. Keywords: Affective user modeling; Affective student modeling; Self-efficacy;
Intelligent tutoring systems; Inductive learning; Human-computer interaction | |||
| Diagnosing and acting on student affect: the tutor's perspective | | BIBAK | Full-Text | 125-173 | |
| KaÅ>ka Porayska-Pomsta; Manolis Mavrikis | |||
| In this paper we explore human tutors' inferences in relation to learners'
affective states and the relationship between those inferences and the actions
that tutors take as their consequence. At the core of the investigations
presented in this paper lie fundamental questions associated with the role of
affective considerations in computer-mediated educational interactions. Theory
of linguistic politeness is used as the basis for determining the contextual
factors relevant to human tutors's actions, with special attention being
dedicated to learner affective states. A study was designed to determine what
affective states of the learners are relevant to tutoring mathematics and to
identify the mechanisms used by tutors to predict such states. Logs of
tutor-student dialogues were recorded along with contextual factors taken into
consideration by tutors in relation to their specific tutorial dialogue moves.
The logs were annotated in order to determine the types and range of student
and tutor actions. Machine learning techniques were then applied to those
actions to predict the values of three factors: student confidence, interest
and effort. Whilst due to limited size and sparsity of data the results are not
conclusive, they are very valuable as the basis for empirically derived
hypotheses to be tested in further studies. The potential implications of the
hypotheses, if they were confirmed by further studies, are discussed in
relation to the impact of tutor's ability to diagnose student affect on the
nature of computer-mediated tutorial interactions. Keywords: Situation modelling; Affect; Machine learning; Tutor feedback; Empirically
based hypotheses generation; Computer based tutoring; Computer mediated
communication | |||
| Private emotions versus social interaction: a data-driven approach towards analysing emotion in speech | | BIBAK | Full-Text | 175-206 | |
| Anton Batliner; Stefan Steidl | |||
| The 'traditional' first two dimensions in emotion research are VALENCE and
AROUSAL. Normally, they are obtained by using elicited, acted data. In this
paper, we use realistic, spontaneous speech data from our 'AIBO' corpus
(human-robot communication, children interacting with Sony's AIBO robot). The
recordings were done in a Wizard-of-Oz scenario: the children believed that
AIBO obeys their commands; in fact, AIBO followed a fixed script and often
disobeyed. Five labellers annotated each word as belonging to one of eleven
emotion-related states; seven of these states which occurred frequently enough
are dealt with in this paper. The confusion matrices of these labels were used
in a Non-Metrical Multi-dimensional Scaling to display two dimensions; the
first we interpret as VALENCE, the second, however, not as AROUSAL but as
INTERACTION, i.e., addressing oneself (angry, joyful) or the communication
partner (motherese, reprimanding). We show that it depends on the specifity of
the scenario and on the subjects' conceptualizations whether this new dimension
can be observed, and discuss impacts on the practice of labelling and
processing emotional data. Two-dimensional solutions based on acoustic and
linguistic features that were used for automatic classification of these
emotional states are interpreted along the same lines. Keywords: Emotion; Speech; Dimensions; Categories; Annotation; Data-driven;
Non-metrical multi-dimensional scaling | |||
| Entertainment capture through heart rate activity in physical interactive playgrounds | | BIBAK | Full-Text | 207-243 | |
| Georgios N. Yannakakis; John Hallam | |||
| An approach for capturing and modeling individual entertainment ("fun")
preferences is applied to users of the innovative Playware playground, an
interactive physical playground inspired by computer games, in this study. The
goal is to construct, using representative statistics computed from children's
physiological signals, an estimator of the degree to which games provided by
the playground engage the players. For this purpose children's heart rate (HR)
signals, and their expressed preferences of how much "fun" particular game
variants are, are obtained from experiments using games implemented on the
Playware playground. A comprehensive statistical analysis shows that children's
reported entertainment preferences correlate well with specific features of the
HR signal. Neuro-evolution techniques combined with feature set selection
methods permit the construction of user models that predict reported
entertainment preferences given HR features. These models are expressed as
artificial neural networks and are demonstrated and evaluated on two Playware
games and two control tasks requiring physical activity. The best network is
able to correctly match expressed preferences in 64% of cases on previously
unseen data (p-value 6*10{sup:-5)). The generality of the methodology, its
limitations, its usability as a real-time feedback mechanism for entertainment
augmentation and as a validation tool are discussed. Keywords: Entertainment modeling; Biosignals; Intelligent interactive playgrounds;
Mixed-reality games; Artificial neural networks | |||
| Mediation of user models for enhanced personalization in recommender systems | | BIBAK | Full-Text | 245-286 | |
| Shlomo Berkovsky; Tsvi Kuflik | |||
| Provision of personalized recommendations to users requires accurate
modeling of their interests and needs. This work proposes a general framework
and specific methodologies for enhancing the accuracy of user modeling in
recommender systems by importing and integrating data collected by other
recommender systems. Such a process is defined as user models mediation. The
work discusses the details of such a generic user modeling mediation framework.
It provides a generic user modeling data representation model, demonstrates its
compatibility with existing recommendation techniques, and discusses the
general steps of the mediation. Specifically, four major types of mediation are
presented: cross-user, cross-item, cross-context, and cross-representation.
Finally, the work reports the application of the mediation framework and
illustrates it with practical mediation scenarios. Evaluations of these
scenarios demonstrate the potential benefits of user modeling data mediation,
as in certain conditions it allows improving the quality of the recommendations
provided to the users. Keywords: Recommender systems; Ubiquitous user modeling; Mediation of user modeling
data | |||
| Developing a generalizable detector of when students game the system | | BIBAK | Full-Text | 287-314 | |
| Ryan S. J. d. Baker; Albert T. Corbett | |||
| Some students, when working in interactive learning environments, attempt to
"game the system", attempting to succeed in the environment by exploiting
properties of the system rather than by learning the material and trying to use
that knowledge to answer correctly. In this paper, we present a system that can
accurately detect whether a student is gaming the system, within a Cognitive
Tutor mathematics curricula. Our detector also distinguishes between two
distinct types of gaming which are associated with different learning outcomes.
We explore this detector's generalizability, and find that it transfers
successfully to both new students and new tutor lessons. Keywords: Gaming the system; Latent response models; Cognitive tutors; Behavior
detection; Machine learning; Generalizable models; Student modeling;
Interactive learning environments | |||
| MUSIPER: a system for modeling music similarity perception based on objective feature subset selection | | BIBAK | Full-Text | 315-348 | |
| Dionysios N. Sotiropoulos | |||
| We explore the use of objective audio signal features to model the
individualized (subjective) perception of similarity between music files. We
present MUSIPER, a content-based music retrieval system which constructs music
similarity perception models of its users by associating different music
similarity measures to different users. Specifically, a user-supplied relevance
feedback procedure and related neural network-based incremental learning allows
the system to determine which subset of a set of objective features
approximates more accurately the subjective music similarity perception of a
specific user. Our implementation and evaluation of MUSIPER verifies the
relation between subsets of objective features and individualized music
similarity perception and exhibits significant improvement in individualized
perceived similarity in subsequent music retrievals. Keywords: Music similarity perception; Relevance feedback; User model; User driven
feature selection; Individualization; Content-based retrieval | |||
| A multifactor approach to student model evaluation | | BIBAK | Full-Text | 349-382 | |
| Michael V. Yudelson; Olga P. Medvedeva | |||
| Creating student models for Intelligent Tutoring Systems (ITS) in novel
domains is often a difficult task. In this study, we outline a multifactor
approach to evaluating models that we developed in order to select an
appropriate student model for our medical ITS. The combination of areas under
the receiver-operator and precision-recall curves, with residual analysis,
proved to be a useful and valid method for model selection. We improved on
Bayesian Knowledge Tracing with models that treat help differently from
mistakes, model all attempts, differentiate skill classes, and model
forgetting. We discuss both the methodology we used and the insights we derived
regarding student modeling in this novel domain. Keywords: Student modeling; Intelligent tutoring systems; Knowledge Tracing;
Methodology; Decision theory; Model evaluation; Model selection; Intelligent
medical training systems; Machine learning; Probabilistic models; Bayesian
models; Hidden Markov Models | |||
| Preface | | BIB | Full-Text | 383-387 | |
| Liliana Ardissono; Daniela Petrelli | |||
| LISTEN: a user-adaptive audio-augmented museum guide | | BIBAK | Full-Text | 389-416 | |
| Andreas Zimmermann; Andreas Lorenz | |||
| Modern personalized information systems have been proven to support the user
with information at the appropriate level and in the appropriate form. In
specific environments like museums and exhibitions, focusing on the control of
such a system is contradictory to establishing a relationship with the
artifacts and exhibits. Preferably, the technology becomes invisible to the
user and the physical reality becomes the interface to an additional virtual
layer: by naturally moving in the space and/or manipulating physical objects in
our surroundings the user will access information and operate the virtual
layer. The LISTEN project is an attempt to make use of the inherent "everyday"
integration of aural and visual perception, developing a tailored, immersive
audio-augmented environment for the visitors of art exhibitions. The challenge
of the LISTEN project is to provide a personalized immersive augmented
environment, an aim which goes beyond the guiding purpose. The visitors of the
museum implicitly interact with the system because the audio presentation is
adapted to the users' contexts (e.g. interests, preferences, motion, etc.),
providing an intelligent audio-based environment. This article describes the
realization and user evaluation of the LISTEN system focusing on the
personalization component. As this system has been installed at the Kunstmuseum
Bonn in the context of an exhibition comprising artworks of the painter August
Macke, a detailed evaluation could be conducted. Keywords: Audio information systems; User-adaptive systems; Context-awareness; User
modeling; Personalization; Personalized audio environments; Acoustic
information spaces; Motion styles; User tracking | |||
| A stroll with Carletto: adaptation in drama-based tours with virtual characters | | BIBAK | Full-Text | 417-453 | |
| Rossana Damiano; Cristina Gena | |||
| In this paper, we present an application for character-based guided tours on
mobile devices. The application is based on the Dramatour methodology for
information presentation, which incorporates a dramatic attitude in
character-based presentations. The application has been developed for a
historical site and is based on a virtual character, "Carletto", a spider with
an anthropomorphic aspect, who engages in a dramatized presentation of the
site. Content items are delivered in a location-aware fashion, relying on a
wireless network infrastructure, with visitors who can stroll freely. The
selection of contents keeps track of user location and of the interaction
history, in order to deliver the appropriate type and quantity of informative
items, and to manage the given/new distinction in discourse. The communicative
strategy of the character is designed to keep it believable along the
interaction with the user, while enforcing dramatization effects. The design of
the communicative strategy relies on the fact that the units of the
presentation are tagged with metadata concerning their content and
communicative function. The description of the application is accompanied by an
evaluation study based on a sample of about 300 visitors, carried out in April
2006, when the installation was open to the public for 1 week. Keywords: Interactive drama; Virtual characters; Cultural heritage; Mobile computing;
Adaptive applications | |||
| The effects of transparency on trust in and acceptance of a content-based art recommender | | BIBAK | Full-Text | 455-496 | |
| Henriette Cramer; Vanessa Evers | |||
| The increasing availability of (digital) cultural heritage artefacts offers
great potential for increased access to art content, but also necessitates
tools to help users deal with such abundance of information. User-adaptive art
recommender systems aim to present their users with art content tailored to
their interests. These systems try to adapt to the user based on feedback from
the user on which artworks he or she finds interesting. Users need to be able
to depend on the system to competently adapt to their feedback and find the
artworks that are most interesting to them. This paper investigates the
influence of transparency on user trust in and acceptance of content-based
recommender systems. A between-subject experiment (N=60) evaluated interaction
with three versions of a content-based art recommender in the cultural heritage
domain. This recommender system provides users with artworks that are of
interest to them, based on their ratings of other artworks. Version 1 was not
transparent, version 2 explained to the user why a recommendation had been made
and version 3 showed a rating of how certain the system was that a
recommendation would be of interest to the user. Results show that explaining
to the user why a recommendation was made increased acceptance of the
recommendations. Trust in the system itself was not improved by transparency.
Showing how certain the system was of a recommendation did not influence trust
and acceptance. A number of guidelines for design of recommender systems in the
cultural heritage domain have been derived from the study's results. Keywords: User-adaptivity; Human-computer interaction; Recommender systems;
Transparency; Trust; Acceptance; Cultural heritage | |||
| Tag-based user modeling for social multi-device adaptive guides | | BIBAK | Full-Text | 497-538 | |
| Francesca Carmagnola; Federica Cena | |||
| This paper aims to demonstrate that the principles of adaptation and user
modeling, especially social annotation, can be integrated fruitfully with those
of the web 2.0 paradigm and thereby enhance in the domain of cultural heritage.
We propose a framework for improving recommender systems through exploiting the
users tagging activity. We maintain that web 2.0's participative features can
be exploited by adaptive web-based systems in order to enrich and extend the
user model, improve social navigation and enrich information from a bottom-up
perspective. Thus our approach stresses social annotation as a new and powerful
kind of feedback and as a way to infer knowledge about users. The prototype
implementation of our framework in the domain of cultural heritage is named
iCITY. It is serving to demonstrate the validity of our approach and to
highlight the benefits of this approach specifically for cultural heritage.
iCITY is an adaptive, social, multi-device recommender guide that provides
information about the cultural resources and events promoting the cultural
heritage in the city of Torino. Our paper first describes this system and then
discusses the results of a set of evaluations that were carried out at
different stages of the systems development and aimed at validating the
framework and implementation of this specific prototype. In particular, we
carried out a heuristic evaluation and two sets of usability tests, aimed at
checking the usability of the user interface, specifically of the adaptive
behavior of the system. Moreover, we conducted evaluations aimed at
investigating the role of tags in the definition of the user model and the
impact of tags on the accuracy of recommendations. Our results are encouraging. Keywords: Web 2.0; Tag; Annotation; User model; Recommendations; Cultural events | |||