Enhanced sentiment analysis And detection through Intelligent chatbot

Abstract

The text represents information s in written or typed form. The text mining extract meaningful information from sentences. The knowledge obtain from different sources such as news articles, comments on social media, customer reviews about product or services, reviews on medical products and much more. The reviews of such products and services are overwhelming to be handled manually by humans. The reviews of such products and service grows exponentially with time. Hence, the sentiment of text automatically analyze by text mining algorithms. The text-mining algorithm classifies reviews to evaluate sentiments. The text mining analyses sentences from different resources to analyze trend, sentiment polarity and linguistic expression. The text mining evaluates similarity between sentences and their meaning. The text mining extracts information from text with respect to specific attributes. The sentences of reviews about products, interaction among different persons have noise such as informal and personnel texts. newlineThe initial process of text mining, removes noisy text from sentences. The morphological analysis determines the different parts of speech in sentence. In chat communication between customer and chat bots, the chat bots analyze semantic features of text to determine customer underlying intention, emotion and sentiment. Furthermore, the aspect based sentiment analysis reviews the sentences for sentiment polarity. The sentiment of text analyze with different algorithms such as Support Vector Machine (SVM), Multi class Support Vector Machine (MSVM) and Minimum Spanning Tree (MST) and Cuckoo Search Optimization algorithm (CSO). newline