Unstructured Social Media Data Mining System Based on Emotional Database and Unstructured Information Management Architecture Framework

We're entering a new age of cloud and big data generated by social media and networks. Social media has become a popular tool that allows people to communicate and share information and the volume of data generated by social media continues to grow at a staggering rate in these days. Social media's influence continues to grow and analyzing social media data is becoming an important issue for research and development and service to learn about what people think and how people feel. Social media data has properties different from traditional data. Thus, traditional data mining methods are not suitable for social media data and new data mining methods are needed to analyze them. In this paper, we develop a new social data mining system which consists of an emotional database and UIMA (Unstructured Information Management Architecture) analysis system. To build emotional database, we collect various unstructured data from social media and classify sentiment texts into eight emotions. UIMA analysis system performs a structure analysis on the social media data related to the keywords user type using the emotional database and provides the statistical analysis result with graphs and numerical values in real time. We perform our experiments on tweet data from twitter to identify and predict sentiment on the topic user select.