Method of optimizing the dimensional features in sentiment analysis

Abstract

The scope to express opinion by the end-user or customer about a product, service, administrative decisions, or events to the world is phenomenal, which is due to the internet access feasibility and the ease of using social web. The opportunity to identify the divergence of the opinion about the corresponding entity (product, service, decision, or event) enables to conclude the opportunity to stabilize, revise, improve, or withdrawal of that entity in regard to amaze the target audience. However, the quantity of the opinions about an entity elevates the difficulty of process to identify the divergence of the opinion using human resource. Hence, the computer-aided methods are the crux to find the divergence of the opinions. Nevertheless, the projection of opinion is not mandate to be in similar passion. Hence, the divergence in nature of expressing the opinion is substantial due to the ambiguity in natural languages and the dimensions of opinion expression feasibilities. In regard to this argument, the design of computer-aided methods to identify the opinion scope is intended in recent past contributions of the research contributions related to natural language processing, which are still inadequate. In order to this argument, the contribution of this manuscript ‘Method of Optimizing the Dimensional Features (MODF)in Sentiment Analysis’, this is a machine learning technique that enables to deal the multiple dimensions of the expressions of the opinions to train the target classifier. The experimental study portraying the significance and scalability of the proposal scaled against the other contemporary models.