Within the Canon of Sherlock Holmes : a python-led study on gender differences in speech

• Main Author: Chan, Alvina Huang Wern
• Other Authors: Francis Bond
• Format: Final Year Project
• Language: English
• Published: 2019
• Subjects: DRNTU::Humanities::Linguistics
• Online Access: http://hdl.handle.net/10356/76528
• Institution: Nanyang Technological University

Gender-related linguistic differences have long been an interest of study. However, the literature reveals that many studies are lacking in terms of objective and standardized methodological approaches. It appears that there is still a lack of consensus among researchers regarding a comprehensive picture of gender differences in language. In a bid to provide a fresh quantitative tool that addresses the current research gaps, this project has designed a Python framework that allows a) the extraction and identification of utterances from a corpus of text for b) analysis in accordance to linguistic categories previously identified to be significant features of differentiation such as parts of speech (POS), sentence length, hedges etc. To test it, the program is used to explore the extent to which male and female speech differ in a data sample of the Canon of Sherlock Holmes (henceforth referred to as the Canon) corpora derived from the NTU-MC (Bond et al., 2013) to demonstrate its range of uses and discuss its potential for use in future research regarding gender-related linguistic differences and beyond in other social science fields.