Sentiment Analysis on Social Media for Emotion Classification
Dilesh Tanna, Manasi Dudhane, Amrut Sardar, Kiran Deshpande, Neha Deshmukh
Abstract
Social media consists of various kinds of emotions and sentiments of its users in the form of electronic media. To analyze the reactions or sentiments of the users on a certain post is also a challenging task. Our project aims to automate this task of analyzing the reactions and the posts and generate a report based on the outcome. A custom social media platform is proposed that would give the ability to the users to perform activities such as post, like, comment, and share. The users can be enrolled in different groups such as business or university. This platform would have the ability to perform sentiment analysis on all the activities of the users in a group and create a report according to their reactions and their posts on the platform. Every activity would be rated based on the sentiments behind it on a scale of -1 and +1, 0 is a neutral sentiment. The appropriate admin would receive the reports which can be used for further actions. The analysis would be performed on various factors such as the response of other users on a post. According to the user's posts, its home page would display other relevant posts. The analysis would facilitate the decision-making process for the admins (around other activities on the platform) and also help detect any need to give special attention to any user in a group, such as a student coping up with depression. Also, the proposed platform can be used to share content on other social media platforms as well. As a result, the users would have a single platform with the ability to do a lot more than any existing social media platform facilitates.