This is a real world project on Sentiment Analysis which takes in text input from the user and predicts if the sentiment of the text is positive or negative.
The following steps were followed for completing this project:
1. Gathering of data: The amazon reviews dataset from Kaggle was used for this project.
2. Preprocessing of data: - Lower casing the text - Expanding contractions - Removing punctuations and special characters - Removing stopwords - Tokenization - Lemmatization
3. Approach to Sentiment Analysis: - TFIDF Vectorizer - Support Vector Machine Model - Evaluation of model using Accuracy Score, Confusion Matrix, and Classification Report
4. Deployment of Model:
- Creating a web application using Streamlit
- Deploying it using Heroku Cloud Service
the link to the website (hosted with Heroku And Streamlit): https://sentiment-analysis-major-proj.herokuapp.com/