A Quantitative Analysis of Lexical Differences Between Genders in Telephone Conversations

Proceedings of the 43rd Annual Meeting of the ACL, pages 435–442, Ann Arbor, June 2005. c⃝2005 Association for Computational Linguistics A Quantitative Analysis of Lexical Differences Between Genders in Telephone Conversations Constantinos Boulis Department of Electrical Engineering University of Washington Seattle, 98195 boulis@ee.washington.edu Mari Ostendorf Department of Electrical Engineering University of Washington Seattle, 98195 mo@ee.washington.edu Abstract In this work, we provide an empiri- cal analysis of differences in word use between genders in telephone conversa- tions, which complements the consid- erable body of work in sociolinguistics concerned with gender linguistic differ- ences. Experiments are performed on a large speech corpus of roughly 12000 con- versations. We employ machine learn- ing techniques to automatically catego- rize the gender of each speaker given only the transcript of his/her speech, achiev- ing 92% accuracy. An analysis of the most characteristic words for each gender is also presented. Experiments reveal that the gender of one conversation side influ- ences lexical use of the other side. A sur- prising result is that we were able to clas- sify male-only vs. female-only conversa- tions with almost perfect accuracy. 1 Introduction Linguistic and prosodic differences between gen- ders in American English have been studied for decades. The interest in analyzing the gender lin- guistic differences is two-fold. From the scientific perspective, it will increase our understanding of language production. From the engineering perspective, it can help improve the performance of a number of natural language processing tasks, such as text classification, machine translation or automatic speech recognition by training better lan- guage models. Traditionally, these differences have been investigated in the fields of sociolinguistics and psycholinguistics, see for example (Coates, 1997), (Eckert and McConnell-Ginet, 2003) or http://www.ling.lancs.ac.uk/groups/gal/genre.htm for a comprehensive bibliography on language and gender. Sociolinguists have approached the issue from a mostly non-computational perspective using relatively small and very focused data collections. Recently, the work of (Koppel et al., 2002) has used computational methods to characterize the differences between genders in written text, such as literary books. A number of monologues have been analyzed in (Singh, 2001) in terms of lexical richness using multivariate analysis techniques. The question of gender linguistic differences shares a number of issues with stylometry and author/speaker attribution research (Stamatatos et al., 2000), (Doddington, 2001), but novel issues emerge with analysis of conversational speech, such as studying the interaction of genders. In this work, we focus on lexical differences be- tween genders on telephone conversations and use machine learning techniques applied on text catego- rization and feature selection to characterize these differences. Therefore our conclusions are entirely data-driven. We use a very large corpus created for automatic speech recognition - the Fisher corpus de- scribed in (Cieri et al., 2004). The Fisher corpus is annotated with the gender of each speaker making it an ideal resource to study not only the character- istics of individual genders but also of gender pairs in spontaneous, conversational speech. The size and 435 scope of the Fisher corpus is such that robust results can be derived …

Content truncated to 500 words