EVALUATION OF UTTERANCES BASED ON CAUSAL KNOWLEDGE RETRIEVED FROM BLOGS

In this paper, we describe the effectiveness of utterance generation using causal knowledge for a dialogue system. Recently, there has been a variety of research on non-taskoriented dialogue systems; however, an effective approach has not yet been developed. One of the most important reasons for this is that non-task-oriented dialogue systems lack common sense knowledge, which is not in their databases. As the first step towards solving this problem, we concentrated on causal knowledge containing reasons and effects, which can provide unwritten meanings for utterance understanding and generating modules. In this paper we investigated how an utterance generated with knowledge related to user input can improve an existing conversational system. Experiment results show that utterance generation using causal knowledge can improve a conversational system.

[1]  Kenji Araki,et al.  A Japanese Natural Language Toolset Implementation for ConceptNet , 2010, AAAI Fall Symposium: Commonsense Knowledge.

[2]  Roxana Gîrju,et al.  Automatic Detection of Causal Relations for Question Answering , 2003, ACL 2003.

[3]  Yuji Matsumoto,et al.  A Method for Retrieving a Similar Sentence and Its Application to Speech Translation , 2004 .

[4]  Ryuta Terashima,et al.  An Analysis of 'Distinctive' Utterances in Non-task-oriented Conversational Dialogue , 2007 .

[5]  Kenji Araki,et al.  A Casual Conversation System Using Modality and Word Associations Retrieved from the Web , 2008, EMNLP.

[6]  Masahide Horita,et al.  ASSESSING THE PLAUSIBILITY OF INFERENCE BASED ON AUTOMATED CONSTRUCTION OF CAUSAL NETWORKS USING WEB-MINING , 2006 .

[7]  Yuji Matsumoto,et al.  Acquiring causal knowledge from text using the connective marker tame , 2005, TALIP.

[8]  Eleazar Eskin,et al.  Detecting Text Similarity over Short Passages: Exploring Linguistic Feature Combinations via Machine Learning , 1999, EMNLP.

[9]  Joseph Weizenbaum,et al.  and Machine , 1977 .

[10]  Pawel Dybala,et al.  Developing a Large-Scale Corpus for Natural Language Processing and Emotion Processing Research in Japanese , 2010 .

[11]  Pawel Dybala,et al.  When Your Users Are Not Serious Using Web-based Associations, Affect and Humor for Generating Appropriate Utterances for Inappropriate Input:Using Web-based Associations, Affect and Humor for Generating Appropriate Utterances for Inappropriate Input , 2010 .

[12]  Satoshi Sekine,et al.  Extracting Causal Knowledge Using Clue Phrases and Syntactic Patterns , 2008, PAKM.

[13]  Yuji Matsumoto,et al.  Japanese Dependency Analysis using Cascaded Chunking , 2002, CoNLL.

[14]  Catherine Havasi,et al.  ConceptNet 3 : a Flexible , Multilingual Semantic Network for Common Sense Knowledge , 2007 .