Objective - Worked on dataset which consists of 20,000 reviews and ratings for different hotels and our goal is to examine how travelers are communicating their positive and negative experiences on online platforms and major objective is what are the attributes that travelers are considering while selecting a hotel.
Approach - Data Preprocessing - EDA - Analysis of reviews as positive or negative - Training various models to get attributes considered - Deployment using best model.
Result - Used Affin Sentiment Analysis to classify a review as positive or negative and best model to get attributes considered in the review by customer is SVC (kernel = linear) which gave us best train and test accuracy among all models trained.