Dhruv Mittal

Data Science Intern

AIE Software Pvt Ltd•  May 2021 - July 2021

I was assigned a project titled Classifying Amazon Reviews Based on Customer Ratings Using NLP. Used NLP and ML algorithms to classify each review into their respective class(rating). 1) Performed extraction of important features from the various features given in the dataset and converted the 5-class model to 3-class.2) Created a data pre-processing pipeline and implemented various NLP algorithms steps such as expansion of contractions, removal of stopwords, lemmatization to reduce noise in data as well as to standardize it.3) Transformed data to vectors using Word2Vec word embeddings and balanced the dataset.4) Applied Random Forest Classification model to the dataframe and achieved a 75% accuracy.5) Working with LSTM to improve classification accuracy of the model.