A visual risk identification and early warning research for college net loan based on microblog texts

By analyzing microblog texts, a visual early warning research of college net loans is conducted in this paper to effectively reduce the negative impact of illegal Internet loans (abbreviated as "net loans") on college students' lives. Two calculation models are proposed in our research: One is the Identification Model of Risk Degree (IMRD) based on the security level of microblog texts about net loans. The other is the Calculation Model of Relationship Closeness (CMRC) based on three dimensions: user's relevant relationship, interaction strength, and interest similarity. IMRD is used to identify the risk of microblog net loan texts, and determine whether to early warn or not. CMRC is utilized to acquire a net loan communication map by describing the closeness level between net loan microblog publishers and their followers, and analyzing the possibility of retweeting. A visual monitoring and early warning platform is constructed based on these two models. With this platform, further spread of net loan information can be prevented by analyzing the graph and cutting off key nodes in time. The early warning mechanism of this research can effectively alleviate the negative influence of illegal net loans.