Sentiment Analysis and Emotion Detection from Short Informal Texts

Abstract

Sentiment analysis is an automated process of the analysing the attitude or opinion, etc., by using natural language processing techniques. The sentiment word refers to attitudes, opinion, point of view in the informal texts. Emotion analysis measures the various feelings that consumer s express on textual data. While Sentiment detects positive, negative, or neutral polarity, emotion detection focuses on the feelings or emotions of a human, such as, happy, sad, anger, etc. Aspect-based sentiment analysis is a technique to find out aspect, features, or attributes from given text information and figure out the corresponding sentiment. After figuring out the consumer s actions, it understands the consumer s deeper needs, which helps in gaining the correct approach on the merits and demerits of the product or ads etc. Analyzed deeply people s opinion or emotions from short texts is a challenging task for researchers. Deep emotion is learning emotion by analysing deeply the text information. Execution time is also important for real time analysis. Therefore, the main goal of the thesis is to improve classification accuracy in low execution time of aspect-based sentiment analysis and emotion detection. newline newline