Social and Movie Video Data Analysis for Representing Sentiments based on ML Approaches

Sentiment analysis (SA) is the process of defining and classifying articulated thoughts or emotions in source documents. As per requirements on the different sentiments of given applications such as Twitter, Movie review data, Amazon purchase reviews, Flipkart purchase/customer reviews have been reviewed and implemented using Machine learning Approaches. Presently, a mathematical solution analysis on identifying the positive, negative and neutral with pie-chart modelling and data representations using Tabulated charts has been proposed. This proposed model uses two different datasets from Kaggle websites such as Twitter and Movie data. Each data is clustered and segregated with different set of class for Twitter and Movie Reviews from the each column considerations on the dataset that chosen. This design improvises on the Data analyzation with XGBOOST algorithms for each movie reviews in which data representation is trained with other ensemble methods such as RFC, ADABOOST and ET. The supervised approach from the class objects considered for Positive, Negative scenarios based on Bayes algorithm that provide 94.53% accuracy and also for Decision Tree (Bagging), and ADABOOST, XG-BOOST algorithm has accuracy of 93.2 rest of Ensemble approaches will suffice with same values at 93%.