We investigate whether one can determine from the transcripts of U.S. Congressional floor debates whether the speeches represent support of or opposition to proposed legislation. To address this problem, we exploit the fact that these speeches occur as part of a discussion; this allows us to use sources of information regarding relationships between discourse segments, such as whether a given utterance indicates agreement with the opinion expressed by another. We find that the incorporation of such information yields substantial improvements over classifying speeches in isolation. 1

We describe a probabilistic approach to content selection for meeting summarization. We use skipchain Conditional Random Fields (CRF) to model non-local pragmatic dependencies between paired utterances such as QUESTION-ANSWER that typically appear together in summaries, and show that these models outperform linear-chain CRFs and Bayesian models in the task. We also discuss different approaches for ranking all utterances in a sequence using CRFs. Our best performing system achieves 91.3 % of human performance when evaluated with the Pyramid evaluation metric, which represents a 3.9 % absolute increase compared to our most competitive non-sequential classifier. 1

We present results on addressee identification in four-participants face-to-face meetings using Bayesian Network and Naive Bayes classifiers. First, we investigate how well the addressee of a dialogue act can be predicted based on gaze, utterance and conversational context features. Then, we explore whether information about meeting context can aid classifiers’ performances. Both classifiers perform the best when conversational context and utterance features are combined with speaker’s gaze information. The classifiers show little gain from information about meeting context. 1

This paper presents experiments into the resolution of “you ” in multi-party dialog, dividing this process into two tasks: distinguishing between generic and referential uses; and then, for referential uses, identifying the referred-to addressee(s). On the first task we achieve an accuracy of 75 % on multi-party data. We achieve an accuracy of 47 % on determining the actual identity of the referent. All results are achieved without the use of visual information. 1

This paper describes a summarization system for technical chats and emails on the Linux kernel. To reflect the complexity and sophistication of the discussions, they are clustered according to subtopic structure on the sub-message level, and immediate responding pairs are identified through machine learning methods. A resulting summary consists of one or more mini-summaries, each on a subtopic from the discussion.

This study addresses the problem of automatically detecting decisions in conversational speech. We formulate the problem as classifying decision-making units at two levels of granularity: dialogue acts and topic segments. We conduct an empirical analysis to determine the characteristic features of decision-making dialogue acts, and train MaxEnt models using these features for the classification tasks. We find that models that combine lexical, prosodic, contextual and topical features yield the best results on both tasks, achieving 72 % and 86 % precision, respectively. The study also provides a quantitative analysis of the relative importance of the feature types. 1

We describe a process for automatically detecting decision-making sub-dialogues in transcripts of multi-party, human-human meetings. Extending our previous work on action item identification, we propose a structured approach that takes into account the different roles utterances play in the decisionmaking process. We show that this structured approach outperforms the accuracy achieved by existing decision detection systems based on flat annotations, while enabling the extraction of more fine-grained information that can be used for summarization and reporting. 1

We investigate the combination of several sources of information for the purpose of subjectivity recognition and polarity classification in meetings. We focus on features from two modalities, transcribed words and acoustics, and we compare the performance of three different textual representations: words, characters, and phonemes. Our experiments show that character-level features outperform wordlevel features for these tasks, and that a careful fusion of all features yields the best performance. 1 1

In this paper, we present approaches used in text summarization, showing how they can be adapted for speech summarization and where they fall short. Informal style and apparent lack of structure in speech mean that the typical approaches used for text summarization must be extended for use with speech. We illustrate how features derived from speech can help determine summary content within two ongoing summarization projects at Columbia University. 1.

We present an annotation scheme for emotionally relevant behavior at the speaker contribution level in multiparty conversation. The scheme was applied to a large, publicly available meeting corpus by three annotators, and subsequently labeled with emotional valence. We report inter-labeler agreement statistics for the two schemes, and explore the correlation between speaker valence and behavior, as well as that between speaker valence and the previous speaker’s behavior. Our analyses show that the co-occurrence of certain behaviors and valence classes significantly deviates from what is to be expected by chance; in isolated cases, behaviors are predictive of valence. 1.