424 Hits in 11.6 sec

Contrasting the interaction structure of an email and a telephone corpus

Jun Hu, Rebecca J. Passonneau, Owen Rambow
2009 Proceedings of the SIGDIAL 2009 Conference on The 10th Annual Meeting of the Special Interest Group on Discourse and Dialogue - SIGDIAL '09   unpublished
We present a dialogue annotation scheme for both spoken and written interaction, and use it in a telephone transaction corpus and an email corpus.  ...  We train classifiers, comparing regular SVM and structured SVM against a heuristic baseline. We provide a novel application of structured SVM to predicting relations between instance pairs.  ...  We then use three algorithms, two of which use machine learning (including a novel approach to using Structured SVM), to predict labels and links (a generalization of adjacency pairs) on unseen data.  ... 
doi:10.3115/1708376.1708429 fatcat:pgvxnjmegrdcbjccrglk63ck5i

Schemas and the frequency/acceptability mismatch: Corpus distribution predicts sentence judgments

Susanne Flach
2020 Cognitive Linguistics  
While a formal model makes no prediction with regard to gradient acceptability of bare ("grammatical") go/come-v, the usage-based view assumes that acceptability is a function of compatibility with an  ...  AbstractA tight connection between competence and performance is a central tenet of the usage-based model.  ...  He uses this approach to model the frequency distribution of a set of morphological forms in a corpus.  ... 
doi:10.1515/cog-2020-2040 fatcat:mn2ygrvqijbkzmedmf7565aoya

Human/Human Conversation Understanding [chapter]

Gokhan Tur, Dilek Hakkani-Tür
2011 Spoken Language Understanding  
We also try to point the bridge studies using human/human conversations for building better human/machine conversational systems or using the approaches for human/machine understanding for better human  ...  The AMI (Augmented Multi-party Interaction) Consortium (AMI n.d.) and DARPAfunded CALO (Cognitive Assistant that Learns and Organizes) (DARPA Cognitive Agent that Learns and Organizes (CALO) Project n.d  ...  This corpus has first been designed to complement the TIMIT corpus (Zue et al. 1990) , which was read speech, and the ATIS corpus (Price 1990) , which was human/machine interactions.  ... 
doi:10.1002/9781119992691.ch9 fatcat:2quzefll5re2lp2ytbasw6nthu

Influence of contextual information in emotion annotation for spoken dialogue systems

Zoraida Callejas, Ramón López-Cózar
2008 Speech Communication  
We have carried out experiments with a corpus extracted from real human interactions with a spoken dialogue system.  ...  Results show that the performance of non-expert human annotators and machine-learned classifications are both affected by contextual information.  ...  Acknowledgements The authors would like to thank the reviewers for their knowledgeable comments and helpful suggestions to improve this paper.  ... 
doi:10.1016/j.specom.2008.01.001 fatcat:3jpm2avtejfltelqj5sjjb2fkq

Automatic analysis of multiparty meetings

2011 Sadhana (Bangalore)  
We discuss the capture and annotation of the AMI meeting corpus, the development of a meeting speech recognition system, and systems for the automatic segmentation, summarisation and social processing  ...  This is a challenging task since the meetings consist of spontaneous and conversational interactions between a number of participants: it is a multimodal, multiparty, multistream problem.  ...  The corpus has been heavily annotated by hand, primarily to provide ground truth for machine learning.  ... 
doi:10.1007/s12046-011-0051-3 fatcat:mdhaduibcnck3b3yi763snppdm

Interaction Mining: The New Frontier of Customer Interaction Analytics [chapter]

Vincenzo Pallotta, Rodolfo Delmonte
2013 Studies in Computational Intelligence  
and key performance indicators (KPIs) beyond the standard approach.  ...  In this paper, we present our solution for pragmatic analysis of call center conversations in order to provide useful insights for enhancing Call Center Analytics to a level that will enable new metrics  ...  These interactions can happen over multiple different channels including telephone, instant messaging, email, web forms, etc.  ... 
doi:10.1007/978-3-642-31546-6_6 fatcat:yqdiy52bfzd2flaqvyrpocht6e

Survey on Evaluation Methods for Dialogue Systems [article]

Jan Deriu, Alvaro Rodrigo, Arantxa Otegi, Guillermo Echegoyen, Sophie Rosset, Eneko Agirre, Mark Cieliebak
2019 arXiv   pre-print
In this paper we survey the methods and concepts developed for the evaluation of dialogue systems. Evaluation is a crucial part during the development process.  ...  Often, dialogue systems are evaluated by means of human evaluations and questionnaires. However, this tends to be very cost and time intensive.  ...  We would also like to thank our anonymous reviewers for their valuable reviews that helped improve the quality of this survey.  ... 
arXiv:1905.04071v1 fatcat:kyeepr7zpzgivhqdkjzfpskv3i

Discovering habits of effective online support group chatrooms

Elijah Mayfield, Miaomiao Wen, Mitch Golant, Carolyn Penstein Rosé
2012 Proceedings of the 17th ACM international conference on Supporting group work - GROUP '12  
Through machine learning analysis of these annotations, we demonstrate that while sentiment predicts a user's stress at the beginning of a chat, dynamic views of efficacy are stronger indicators of stress  ...  Typically, demonstrating such claims about social interaction would be approached through the lens of sentiment analysis.  ...  Acknowledgements The research reported here was supported by National Science Foundation grant IIS-0968485.  ... 
doi:10.1145/2389176.2389216 dblp:conf/group/MayfieldWGR12 fatcat:gx3izvb5yfhndbd4mngvm7dgee

Conclusion and Perspectives [chapter]

2014 Multimodal Interactive Systems Management  
The authors and publishers express their thanks to the Ecole polytechnique fédérale de Lausanne (EPFL) for its generous support towards the publication of this book. List of contributors  ...  The AMI Meeting Corpus: A pre-announcement. In Proceedings of the 2nd International Workshop on Machine Learning for Multimodal Interaction(MLMI 2005), pages 28-39, Edinburgh, UK.  ...  methods, i.e. by learning the correspondences between features of an input signal and the desired annotation.  ... 
doi:10.1201/b15535-30 fatcat:pizevl5wrzbinctnvshmftmpiq

Interactive Multimodal Information Management [chapter]

Hervé Bourlard
2014 Multimodal Interactive Systems Management  
Thanks to research in the past decade, a convincing case has been made for the capability of information and communication technology to do justice to one of the richest aspects of human communication:  ...  This book is an attempt to answer the questions: what is required from computer hardware and software to support such activities?  ...  Acknowledgments The contributors to this book are members of the IM2 NCCR. Unless otherwise stated, the research work described in this book was funded by the  ... 
doi:10.1201/b15535-2 fatcat:fsjlqstpfze2xng2o7e6ey6zya

Recognition of Dialogue Acts in Multiparty Meetings Using a Switching DBN

A. Dielmann, S. Renals
2008 IEEE Transactions on Audio, Speech, and Language Processing  
We have carried out experiments on the AMI corpus of multimodal meeting recordings, using both manually transcribed speech, and the output of an automatic speech recogniser, and using different configurations  ...  This generative approach models a set of features, related to lexical content and prosody, and incorporates a weighted interpolated factored language model.  ...  DA labels may incorporate syntactic, semantic and pragmatic factors: in addition to providing information about the structure of a dialogue and the course of a conversation, DAs are also able to capture  ... 
doi:10.1109/tasl.2008.922463 fatcat:smulxi22x5hbjjlpk3i4clgede

Multiparticipant chat analysis: A survey

David C. Uthus, David W. Aha
2013 Artificial Intelligence  
the data needed, and completing and reviewing the collection of information.  ...  Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently  ...  The views and opinions contained in this paper are those of the authors and should not be interpreted as representing the official views or policies, either expressed or implied, of NRL or the DoD.  ... 
doi:10.1016/j.artint.2013.02.004 fatcat:lie3tzwvdnezjpcpny4xd52jrm

Automatic Speech Summarisation: A Scoping Review [article]

Dana Rezazadegan, Shlomo Berkovsky, Juan C. Quiroz, A. Baki Kocaballi, Ying Wang, Liliana Laranjo, Enrico Coiera
2020 arXiv   pre-print
We reviewed a total of 110 papers out of a set of 153 found through a literature search and extracted speech features used, methods, scope, and training corpora.  ...  We also discuss the strengths and weaknesses of these different methods and speech features. Overall, supervised methods (e.g.  ...  Another approach relied on discourse relations, in order to learn a general semantic structure and link the turns in a conversation [21] .  ... 
arXiv:2008.11897v1 fatcat:yer4tskqt5bbpk3kyvcvnlenbq

Thai speech processing technology: A review

Chai Wutiwiwatchai, Sadaoki Furui
2007 Speech Communication  
This paper presents a comprehensive review of Thai speech technology, from its impetus in the early 1960s to 2005.  ...  It has a spelling system using a Thai alphabet, but has no explicit word boundaries, similar to several Asian languages, such as Japanese and Chinese.  ...  The authors would also like to express thanks to anonymous reviewers for their constructive suggestions and comments on the first version of the manuscript.  ... 
doi:10.1016/j.specom.2006.10.004 fatcat:w6ruekbjtzazxaqs5d2xcqvr2q

A Dual-Attention Hierarchical Recurrent Neural Network for Dialogue Act Classification [article]

Ruizhe Li, Chenghua Lin, Matthew Collinson, Xiao Li, Guanyi Chen
2019 arXiv   pre-print
Experimental results show that by modelling topic as an auxiliary task, our model can significantly improve DA classification, yielding better or comparable performance to the state-of-the-art method on  ...  With a novel dual task-specific attention mechanism, our model is able, for utterances, to capture information about both DAs and topics, as well as information about the interactions between them.  ...  Acknowledgment This work is supported by the award made by the UK Engineering and Physical Sciences Research Council (Grant number: EP/P011829/1).  ... 
arXiv:1810.09154v3 fatcat:6c7fdlmocfa7bclvibly7mrbom
« Previous Showing results 1 — 15 out of 424 results