A framework for sentiment analysis in customer product reviews using machine learning

Perceiving and Analysis of textual information created from different online life stages have gotten one of the most necessities in the present Big Data period. The consequence of such investigation helps in numerous vital organizations to increase clear experiences about their plans of action and to in the long run take essential business-situated choices to improve their organizations. In this paper, an endeavor is made to perform a notion examination on the conveyed processing system utilizing the many Machine Learning (ML) models and Hadoop-based Spark programming model. The current methodologies towards slant investigation are restricted to just a couple of brands and their items. In this way, to coordinate the learning capacities with conveyed processing models on huge literary information, we built up the proposed system that prescribes the item to clients as per client's element prerequisites gathered as the enormous printed information. The investigation executed the Gaussian Naive Bayes (GNB) and Random Forest (RF) on the Spark Big Data examination stage to process immense printed information. The trial results have indicated that the two calculations produce better effectiveness over different strategies while handling large opinion datasets.