A Sequential Model for Discourse Segmentation

Entities Mentioned

• Japan
• Tokyo
• Conditional Random Fields
• Hongo
• JapanMitsuru Ishizuka

1. Identifying discourse relations in a text is essential for various tasks in Natural Language Processing, such as automatic text summarization, question-answering, and dialogue generation. The first step of this process is segmenting a text into elementary units. In this paper, we present a novel model of discourse segmentation based on sequential data labeling. Namely, we use Conditional Random Fields to train a discourse segmenter on the RST Discourse Treebank, using a set of lexical and syntactic features. Our system is compared to other statistical and rule-based segmenters, including one based on Support Vector Machines. Experimental results indicate that our ... (Read Full Article)

   Bookmark or Share this article

   
   

   Related Articles

   • Cognition Announces the Commercial Release of Its Semantic Natural ...
   • also categorized in Semantic
   • Automatic Identification of Legal Terms in Czech Law Texts
   • also published in SpringerLink Home
   • A Linguistic Interpretation of the OCC Emotion Model for Affect Sensing from Text
   • also mentions JapanMitsuru Ishizuka
   • Cognition Announces Natural Language Processing Technology
   • also categorized in Semantic
   • Introducing Context and Reasoning in Visual Content Analysis: An Ontology-Based Framework
   • also published in SpringerLink Home

   • 2006 ACL Lifetime Achievement Award
   • also mentions Tokyo
   • Gram-Free Synonym Extraction Via Suffix Arrays
   • also mentions Hongo
   • Taxonomy-based partitioning of the Gene Ontology.
   • also categorized in Semantic
   • Cognition Announces the Commercial Release of Its Semantic Natural ...
   • also categorized in Semantic
   • 2003 ACL Lifetime Achievement Award
   • also mentions Japan

Login to comment.