Techniques for topic detection based processing in spoken dialog systems

Balchandran, Rajesh; Rachevsky, Leonid; Ramabhadran, Bhuvana; Novák, Miroslav

doi:10.21437/interspeech.2010-39

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These aspects are discussed for text processing in (Eisenstein and Barzilay, 2008) together with a critical review on lexical cohesion. In (Balchandran et al, 2010) techniques for topic detection have been proposed for obtaining topic specific language models (lms) used for re-scoring asr results.…”

Section: Related Workmentioning

confidence: 99%

Multiple topic identification in human/human conversations

Bost

Senay

El-Bèze

et al. 2015

Computer Speech & Language

View full text Add to dashboard Cite

This paper deals with the automatic analysis of conversations between a customer and an agent in a call centre of a customer care service. The purpose of the analysis is to hypothesize themes about problems and complaints discussed in the conversation. Themes are defined by the application documentation topics. A conversation may contain mentions that are irrelevant for the application purpose and multiple themes whose mentions may be interleaved portions of a conversation that cannot be well defined. Two methods are proposed for multiple theme hypothesization. One of them is based on a cosine similarity measure using a bag of features extracted from the entire conversation. The other method introduces the concept of thematic density distributed around specific word positions in a conversation. In addition to automatically selected words, word bigrams with possible gaps between successive words are also considered and selected. Experimental results show that the results obtained with the proposed methods outperform the results obtained with support vector machines on the same data. Furthermore, using the theme skeleton of a conversation from which thematic densities are derived, it will be possible to extract components of an automatic conversation report to be used for improving the service performance.

show abstract

Section: Related Workmentioning

confidence: 99%