Enhancing sentiment analysis on social media data

• Main Author: Ruan, Pingcheng
• Other Authors: Li Fang
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/66602
• Institution: Nanyang Technological University

This report summarizes the entire work for my final year project SCE15-0580, Enhancing Sentiment Analysis on Social Media Data. The project extends the work on the patented bilingual sentiment analytical engines ChiEFS and AS-SEA during industrial attachment in the Institute of High Performance Computing (IHPC), A*STAR. This report explains my extensive research on the engine’s improvement from the following three perspectives: Fine-grained analysis of the sentiment including specific emotions, associate the sentiment with the topics and compute a domain-specific word’s polarity. The experiments on the testing micro blogs have shown the acceptable performance on the proposed methods. The report also introduces the web platform developed by me, SentiAgentPlus, which demonstrates the above research work and applies the sentiment analytical technique to real time tweets.