Detecting Non-personal and Spam Users on Geo-tagged Twitter Network
Track: Education and Training
Authors: Diansheng Guo, Chao Chen
Non-personal accounts, spam users and junk tweets pose severe problems to the extraction of meaningful information from tweets or twitter users. In this study, we develop a methodological framework to (1) extract user characteristics based on geographic, graph-based and content-based features of tweets, (2) construct a training data by manually inspecting a large sample of tweets and users, and (3) derive reliable rules for detecting non-personal users with supervised classification methods.