A Novel Elephant Herd Random Forest Machine Learning Approach To Estimate the Sentiment Value of Online Customer Review

Abstract

Nowadays, big data has attracted the attention of whole advanced world with its applications and features. Machine learning models are used to run the online administrations in better manner .The machine learning approaches are turned into a moving field in analyzing enormous data; consequently the accomplishment of online administrations or business depends on the client audits. Nearly the online customer review contains positive, negative and neutral sentiment value. In marketing system and product development fields the sentiment analysis value prediction has important role. In this paper, a novel Elephant Herd Random Forest Machine Learning (EHRFML) approach is proposed to estimate the sentiment value of online customer review. Moreover, the customer review datasets are preprocessed and also unwanted information is removed using machine learning approach. Sequentially, the proposed system outcomes are compared with existing technique in terms of accuracy, precision, recall, aspect term specification and opinion condition etc, and achieved better results by obtaining high accuracy and precision of opinion specification.