Interactive Facebook semantic analyser / Nur Azmina Mohamad Zamani … [et al.]

Facebook has become an important platform for interaction among people at any time and any place. It can be used by anybody for various purposes such as socializing, marketing products, expressing sentiments and discussing work-related issues. The concerns about Facebook comments are taken into considerations since it contains comments in the form of people’s opinions and sentiments where they express feelings as well as convey emotions via texts and emoticons. Based on several studies and observations, there are limited number of computerized systems in measuring people’s sentiment in Malay. In addition, monitoring information becomes difficult since there are too many suggestions and the problems tend to be overlooked which may lead to crime cases. Moreover, sentiment analysis in English is a must for most of the Social Network Sites monitoring tools. Therefore, this project aims to identify the opinion mining and sentiment analysis components for extracting both English and Malay words in Facebook, to cluster the extracted texts into emotions (happy, unhappy and emotionless), and also to develop a prototype to quantify people’s sentiments through Facebook posts. The prototype extracts Facebook comments automatically before the comments are filtered by removing unwanted tags. The words are stored in a database to be analyzed semantically by using Lexicon-based approach. Then, results will display the people’s feedbacks toward particular issues that are beneficial to the Facebook owners in order to make further actions.