Sentiment Analysis of Tweets in Hindi

• Course: CSE4022 Natural Language Processes (VIT)
• Tech Stack: Jupyter Notebook, Python (pandas, sklearn, fastai, numpy, NLTK)

• Sentiment analyzer that can determine sentiment polarity of twitter data set
• Cleaned dataset containing 10,000 tweets, removed stop words and performed vectorization
• Trained different classifiers and used FastAI’s ULMFiT model to extract features and classify the Hindi dataset