Media.Player

Automated natural language processing tool uses machine learning to identify quoted speech in literary narratives

Summary

Kathleen McKeown, Ph.D.

Technology Benefits

Addresses the application of natural language processing to literatureLeverages two aspects of the semantics of quoted speech (dialogue chains and the frequent use of expression verbs) by using a pattern matching algorithm to assign each quote to a syntactic category.Draws upon a corpus of more than 3,000 quotations Achieves 83% accuracy in assigning speakers to quotesPatent information:Patent PendingTech Ventures Reference: IR Proxy66

Technology Application

Sentiment analysis (i.e. opinion mining)Political discourse analysis (debates, speeches, hearings, etc.) Automated literary analysis Social network extraction Automatic text summarizationSpeech recognition and segmentation

Detailed Technology Description

Kathleen McKeown, Ph.D.

*Abstract

None

*Inquiry

Jay HickeyColumbia Technology VenturesTel: (212) 854-8444Email: TechTransfer@columbia.edu

*IR

Proxy66

*Principal Investigator

*Publications

Elson DK, McKeown KR. "Automatic Attribution of Quoted Speech in Literary Narrative." Proceedings of the Twenty-Fourth AAAI Conference on Artificial Intelligence. 2010:1013-1019.Elson and McKeown, Automatic Attribution of Quoted Speech in Literary Narrative, AAAI 2010

Country/Region

USA