Detection of Women Profanity in Twitter

Abstract

AbstractIn recent years, global users have been allowed to meet and participate in discussions on various social media platforms. In particular, online communities have evolved to become significant places for interpersonal communication. In recent times, the identification of offensive language in online content has become a topic of rising concern. It can be observed that abusive or vulgar language is being used both in the form of jokes and hate conversation. And those are humiliating and insulting words or phrases. The hate conversation may be based on gender orientation, caste, race, and sex. Due to a number of factors, the detection of abusive language is more often complicated than one assumes. This is not only a challenging work to automate but also a difficult one for individuals due to the noise in the data and the need for global awareness. In this paper, we develop a collection of comments elucidated for hate speech and propose an application that detects profanity in social media. The pre-processing of the dataset is done and then further visualization of the most commonly used terminology in twitter, and the classified hate speech is displayed using word clouds. The text length of the tweets that are considered to contain offensive words by the CF coders is shown by the histogram.KeywordsOnline social media (OSM)Online hate classifierHate speechOffensive languagePre-processingVisualizing