You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The primary goal of this project is to examine whether the extracted and encoded Tamil multi-word tokens exhibit greater semantic similarity to their respective translations in English or in German. This similarity will be assessed using cosine similarity as a metric.
A complete pipeline for sentiment analysis using Hugging Face Transformers and AWS services. The model can be run on both Streamlit Share Server and AWS (using S3 for storage and EC2 for deployment). This repository covers data preprocessing, model training, evaluation, and accurate sentiment prediction on reviews.