Sinhalese Language Based Hate Speech Detection and Intent Analysis on Tweets

• Researched developing a tool to detect Sinhalese language-based hate speech and analyze the intention of tweets.
• Semantic analysis and data mining techniques used in addressing the problem.
• Technologies used: Python, NLTK, Pandas, NumPy, Scikit-learn, PyCharm.

Text Classification Module

Researched Sinhalese Language Based Short Text Classification
to Improve Information Filtering

Module 2: Extract Sinhala twitter texts, Sinhala text preprocessing and Categorize tweet content considering Subject and Domain .
My contribution for the project was implementing the text preprocessing module of which outcomes are used as inputs for analysis, process of classifying sinhala texts found on Twitter into 5 primary domains(Religion, Racist, Political, Sports, Sexism and Others) and further classifying Sinhala texts into 3 subjective categories(Opinion, News, Emotion)
Purpose of implementing this module is to further analyze hate according to particular domain and subject classes to define severity level of the twitter text content.