A Model of an Automatic Detection of Misogynistic Content in Social Media in a Multilingual Environment

ABSTRACT:

There has been enormous growth in the volume of user content posted to social media such as Twitter, Instagram, Facebook, and YouTube. A major challenge for this media is to monitor and prevent the posting of abusive and misogynistic user content, be this bullying content or other types of abusive text content, especially against women. Research work in this area is ongoing. A particular gap, however, is the detection of content across multiple languages especially vernacular, as the tendency is to focus on major languages such as English and Spanish. This research seeks to address both the detection of misogynistic content and the challenge of doing this while taking account of multiple languages. The research will also investigate the relationship between misogyny and other abusive language scenarios in the context earlier stated. Dataset for this research was gotten from social media platforms through their Application Programming Interface (API). The techniques deployed include traditional machine learning algorithms used in building automated modules for categorizing content, text mining for the parsing and analysis of text, and deep learning for learning transfer across different languages/datasets. The proposed conceptual model will be a huge step towards a more ambitious goal of creating novel mechanisms to detect abusive and misogynistic behaviors on social media platforms.

KEYWORDS: Misogyny, Machine learning, Text classification, Multilingual