Automated Content Analysis: A Sentiment Analysis on Malaysian Government Social Media

Social media has become the current trend in information technology industry that acts as the collective of online communication channels for community-based input, interaction, content-sharing and collaboration, used all around the world. Microblogging platforms such as Facebook and Twitter are fast communication channels for information sharing among worldwide users. An automated content analysis tool is proposed in this paper to help Malaysian legal firms and the Malaysian official leaders understand public sentiment via their comments on official Malaysian government leaders' social media sites such as Twitter and Facebook. In this paper, we explore and apply Semantic Role Labeling (SRL) techniques that generate new methods to filter and classify the content data set, advancing the state of the art in sentiment detection approaches, which currently suffer from lack of accuracy. In particular, semantic role labeling techniques have been little studied for use on degraded social media text, which causes numerous problems for traditional Natural Language Processing (NLP) techniques. This work provides a platform for society, especially Malaysian government Legal Firms, IT Agencies and the public as a whole, to measure the impact of public sentiment over Malaysian government officials for policy making and the future development in Malaysia. The work is novel in its application of SRL techniques to social media, and in the use of SRL for improving the quality of sentiment analysis.