Tell me how you write and I will tell you who you are. Genderlects on network

Identifier:  TFG:1537

Handle: https://hdl.handle.net/20.500.11797/TFG1537

Authors:  Giménez Bedós, Clàudia

Abstract:
In the last decades the Internet has become an information and communication network which reaches a wide audience and which has spread all over the world. Within this communicative context it is of paramount importance to know who hides behind a specific linguistic profile in computer-based communication. In order to establish the features of male and female genderlects, sociolinguistics has been based on oral interactions, but, ¿is traditional classification of male and female language still valid, or should it be updated adapting to this new communication medium? The aim of this project has two sides. On the one hand, we have carried out an analysis just to detect if the traditional characterization of genderlects matches with the linguistic traits which are observed in computer-based communicative exchanges between men and women. On the other hand, we have tried to offer a genderlect characterization which could be useful concerning automatic gender detection tasks. Based on specialized bibliography, this project establishes the traditional stereotype of masculinity and femininity. A grand total of 1,200 posts taken from Twitter involving 40 men and 40 women have been analyzed with the aim of checking whether the stereotype was still valid or not. The data obtained through this analysis have eventually been used as the basis for setting up a list of characteristics which may be relevant when detecting automatically either men or women’s identity on the web.