Transfer cross entropy for fast sociometric inference in longitudinal collections of multi-party conversation

Author

Laskowski, Kornel

Author_Institution

Carnegie Mellon Univ., Pittsburgh, PA, USA

fYear

2012

fDate

25-30 March 2012

Firstpage

2189

Lastpage

2192

Abstract

A means is proposed of extracting participant-pair interaction measures from a binary representation of behavior in multi-party conversation, by leveraging an extension of transfer entropy. The technique allows for the inexpensive construction of sociomatrices, requiring only a minimum of detection technology. It is expected that the method will tractably enable the application of social network analysis to conversational behavior mined from very large collections of unannotated audio.

Keywords

entropy; inference mechanisms; social aspects of automation; binary representation; conversational behavior; detection technology; longitudinal collection; multiparty conversation; participant pair interaction measures; social network analysis; sociomatrices; sociometric inference; transfer cross entropy; unannotated audio; Entropy; Equations; History; Mathematical model; Production; Social network services; Speech; N-gram models; Turn-taking; cross entropy; influence; social network analysis; transfer entropy;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on

Conference_Location

Kyoto

ISSN

1520-6149

Print_ISBN

978-1-4673-0045-2

Electronic_ISBN

1520-6149

Type

conf

DOI

10.1109/ICASSP.2012.6288347

Filename

6288347