A sentiment analysis project focused on classifying customer reviews on Amazon products as positive, negative, or neutral. The project is implemented using the Random Forest algorithm and various NLP techniques in Python.
- Python
- Random Forest
- Numpy
- Pandas
- NLTK
- Scikit-learn
- Matplotlib
- Count Vectors Score: 0.92383
- TF-IDF Word-Level Score: 0.91525
- TF-IDF N-Gram Score: 0.85679
The model was further optimized through Hyperparameter tuning and Cross-validation with GridSearchCV.
This resulted in a 6% improvement in accuracy compared to the baseline model (0.92).