A Novel Artificial Intelligence Inspired Approach for Analysing Sentiments using Neural Network

Abstract

Sentiment Analysis is a text classification system used to analyse the incoming text and determine if the sentiment evolved from the argument text is positive, negative or neutral. It holds specific significance in different areas of Technology and it has varied applications. It combines the power of two most significant fields of Artificial Intelligence, viz. Deep Learning and Natural Language Processing. Deep Learning has evolved as an advanced and efficient technology to target and solve real-world problems with greater accuracy and efficiency. Neural Networks are used to deal with enormous amounts of data without any considerable domain engineering and adapt from its own method of computing. Sentiments within a text can become misleading if the whole sentence is taken as a collection of independent words, since each and every word is connected to other in some sense. This research is used to propose and develop an efficient model that can analyse the sequential sentiments with greater prediction and accuracy. This research also compares the prediction accuracy of the proposed model with other existing novel methods.