Mar 2021 – May 2021
Aims to develop a sentiment analysis model using Recurrent Neural Networks (RNN).
The project includes three components, the Word Embedding module, Lexicon Embedding module, and Bi-directional LSTM / RNN Sequence Model.
Sentiment analysis is contextual mining of text which identifies and extracts subjective information in source material, and helps a business to understand the social sentiment of their brand, product or service while monitoring online conversations.
All sections can be run directly.
In the second code cell in Section 1.1, you need to use the generated link to access your Google Account.
Most of the code cells (for parameter tuning, would cost a lot of time) in Section 3 have been comment out.