Abstract:
As online content continues to grow, so does the spread of hate speech. The use of vast online social communication forums helps the user to express their opinion freely at any time. While the ability to freely express oneself is a human right that should be cherished, inducing and spreading hate towards another group is an abuse of this liberty. The availability of a large amount of data with demographical information such as location, time, and events are helpful to analyze the hidden patterns and understanding spontaneously expressed opinions in the sentiment analysis process which would enable more accurate results. Sentiment Analysis is a technique that is being used abundantly nowadays for customer reviews analysis, popularity analysis of electoral candidates, hate speech detection, and similar applications. This thesis aims to perform a spatiotemporal-based sentiment analysis of hate speech tweets in the Pakistan region. The process starts with the collection of political and religious-oriented demographic tweets through Twitter data API and web scrapper. After necessary text preprocessing the refine dataset will be annotated in three categories (Positive, Neutral, and offensive). The labeled data will be passed to the feature extraction module for relevant feature mining. The extracted feature will be trained on state-of-the-art machine learning and deep learning classifiers to investigate the performance of the proposed model.
