Public service hot issue discovery with binary differential evolution algorithm based on fuzzy system theory

Social media is becoming more and more closely related to the real life. More and more netizens choose to obtain news and publish notice through social networks. Such huge amount of social media information generated by these users contains a lot of information related to hot topics and events. At the same time, problem of information overload has posed a challenge for people to use the information. It has become an important research issue to discover and track hot events and topics automatically from mass social media data. On the one hand, the short, highly noisy and real-time features of the social media data bring challenges to the discovery and tracking methods of traditional hot issues. On the other hand, the social media data contains abundant information of geography, time, and social relations, which brings great convenience to relevant researches. Based on these features of the social media data, this paper makes a deep study on the discovery, extraction, and tracking of hot issues in the social media based on fuzzy system theory and the word vector semantic clustering.