This jupyter notebook is to classify amazon reviews into positive and negative reviews using Natural Language Processing, the following dataset has been used.
The following has been conducted in the notebook:
1- Data Preprocessing
2- Stemming reviews using NLTK library.
3- Transforming text into word2vector using google word2vec, and TF-IDF.
4- Applying different machine learning models to predict positive, negative sentiment.
Best accuracy achieved 0.89 f1 score, with SVM TF-IDF