Project information

• Category: AI, Python
• Model Type: Sequential
• Project Date: 01 March, 2024
• Project Link: Github

This project involved developing a tweet emotion classifier that categorizes tweets into six distinct emotions: sadness, joy, love, anger, fear, and surprise. Leveraging natural language processing techniques, I preprocessed tweet data by tokenizing and padding sequences before training a simple yet effective neural network model featuring an embedding layer and a bidirectional GRU layer with a test accuracy of 93.47%, demonstrating that simplicity can yield robust performance. In addition, I performed an in-depth analysis of the unique vocabulary associated with each emotion, providing valuable insights into the linguistic nuances of social media text. This project highlights my ability to apply machine learning and data visualization techniques to solve real-world problems efficiently.