Emotion Classification in Social Networking Sites Using Lexicon Analysis

Abstract

Social Media is used by practically everybody. It is used as a medium to express freely, the thoughts, opinions, beliefs, behaviours and a lot more. This information in regard of a product, service or anything being used by people can serve a great purpose. It serves the purpose of a feedback mechanism. In terms of business, such feedback may be used to improve the service being provided, upgrade the product already in the market so that better releases can be used. In terms of education, such feedback information can be used for feedback of a course, or a teacher. Hence this project has been thought of. Right now, what has already been developed is the part where we extract data from a social networking site, Twitter, and the part where we try to make something meaningful out of the sentences that have been extracted by tokenizing them and identifying parts of speech. Also discussed in the report are the procedures to go about the above two steps and the future work of the project, which includes comparing this lexicon based approach with a machine learning based approach and then if possible, integrating this project with the Student Feedback System of Jaypee University of Information Technology, Waknaghat.