糖心原创

Project description

Social media such as microblogs (eg Twitter) and networking platforms (eg Facebook) are increasingly used for users to communicate breaking news and connect to each other anytime, from anywhere. Social media web sites contain various types of services and therefore different formats of data, including text, image, video etc are created. Among the various formats of data exchanged in social media, text plays a important role. The volume of textual data in social media is increasing exponentially, thus providing us with numerous opportunities for detecting the occurrence of an event in real-time (eg earthquakes, tsunami).

This project is concerned with developing computational statistics and machine learning methodology for text-analytics. In particular we are interested in detecting the occurrence of major events (eg. major train delay) when in the data (for example, tweets) there is often information about minor events too (eg someone tweeting/complaining because their train was 3 minutes late).