In this paper we study the utility of discourse structure for spoken dialogue performance modeling. We experiment with various ways of exploiting the discourse structure: in isolation, as context information for other factors (correctness and certainty) and through trajectories in the discourse structure hierarchy. Our correlation and PARADISE results show that, while the discourse structure is not useful in isolation, using the discourse structure as context information for other factors or via trajectories produces highly predictive parameters for performance analysis. 1

In this study, we incorporate automatically obtained system/user performance features into machine learning experiments to detect student emotion in computer tutoring dialogs. Our results show a relative improvement of 2.7 % on classification accuracy and 8.08 % on Kappa over using standard lexical, prosodic, sequential, and identification features. This level of improvement is comparable to the performance improvement shown in previous studies by applying dialog acts or lexical-/prosodic-/discourse- level contextual features. Index Terms: emotional speech, emotion detection, spoken dialog systems

Given the growing complexity of tasks that spoken dialogue systems are trying to handle, Reinforcement Learning (RL) has been increasingly used as a way of automatically learning the best policy for a system to make. While most work has focused on generating better policies for a dialogue manager, very little work has been done in using RL to construct a better dialogue state. This paper presents a RL approach for determining what dialogue features are important to a spoken dialogue tutoring system. Our experiments show that incorporating dialogue factors such as dialogue acts, emotion, repeated concepts and performance play a significant role in tutoring and should be taken into account when designing dialogue systems. 1

Recent work in designing spoken dialogue systems has focused on using Reinforcement Learning to automatically learn the best action for a system to take at any point in the dialogue to maximize dialogue success. While policy development is very important, choosing the best features to model the user state is equally important since it impacts the actions a system should make. In this paper, we compare the relative utility of adding three features to a model of user state in the domain of a spoken dialogue tutoring system. In addition, we also look at the effects of these features on what type of a question a tutoring system should ask at any state and compare it with our previous work on using feedback as the system action. 1

Speech recognition problems are a reality in current spoken dialogue systems. In order to better understand these phenomena, we study dependencies between speech recognition problems and several higher level dialogue factors that define our notion of student state: frustration/anger, certainty and correctness. We apply Chi Square (χ2) analysis to a corpus of speech-based computer tutoring dialogues to discover these dependencies both within and across turns. Significant dependencies are combined to produce interesting insights regarding speech recognition problems and to propose new strategies for handling these problems. We also find that tutoring, as a new domain for speech applications, exhibits interesting tradeoffs and new factors to consider for spoken dialogue design. 1

Abstract not found

Abstract. Due to the relatively simple structure of dialogues in previous spoken dialogue systems, discourse structure has seen limited applications in these systems. We investigate the utility of discourse structure for spoken dialogue systems in complex domains (e.g. tutoring). Two types of applications are being pursued: on the system side and on the user side. On the system side, we investigate if the discourse structure information is useful for various spoken dialogue system tasks: performance analysis, characterization of user affect and characterization of speech recognition problems. On the user side, we investigate whether the discourse structure information is useful for users of a spoken dialogue system through a graphical representation of the discourse structure.

Tutorial dialogue has been the subject of increasing attention in recent years, and it has become evident that empirical studies of human-human tutorial dialogue can contribute important insights to the design of computational models of dialogue. This paper reports on a corpus study of human-human tutorial dialogue transpiring in the course of problemsolving in a learning environment for introductory computer science. Analyses suggest that the choice of corrective tutorial strategy makes a significant difference in the outcomes of both student learning gains and selfefficacy gains. The findings reveal that tutorial strategies intended to maximize student motivational outcomes (e.g., self-efficacy gain) may not be the same strategies that maximize cognitive outcomes (i.e., learning gain). In light of recent findings that learner characteristics influence the structure of tutorial dialogue, we explore the importance of understanding the interaction between learner characteristics and tutorial dialogue strategy choice when designing tutorial dialogue systems. 1

The field of intelligent tutoring systems has seen many successes in recent years. A significant remaining challenge is the automatic creation of corpus-based tutorial dialogue management models. This paper reports on early work toward this goal. We identify tutorial dialogue modes in an unsupervised fashion using hidden Markov models (HMMs) trained on input sequences of manually-labeled dialogue acts and adjacency pairs. The two best-fit HMMs are presented and compared with respect to the dialogue structure they suggest; we also discuss potential uses of the methodology for future work.

The field of intelligent tutoring systems has seen many successes in recent years. A significant remaining challenge is the automatic creation of corpus-based tutorial dialogue management models. This paper reports on early work toward this goal. We identify tutorial dialogue modes in an unsupervised fashion using hidden Markov models (HMMs) trained on input sequences of manually-labeled dialogue acts and adjacency pairs. The two best-fit HMMs are presented and compared with respect to the dialogue structure they suggest; we also discuss potential uses of the methodology for future work.