Selecting transfer entropy thresholds for influence network prediction

McKenney, Dave; White, Tony

doi:10.1007/s13278-017-0421-x

Cited by 2 publications

(1 citation statement)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, the strength of social network relations is also related to the formation of social relations. McKenney, Dave, White, Tony (2017) [11] mainly contributed to the method of selecting the transfer entropy threshold, which determines whether a link should be considered. It is influential.…”

Section: Introductionmentioning

confidence: 99%

Research on MIL Search Method for P2P Borrowing Loss Clients Based on Social Network Perspective

Pang¹,

Yang²

2018

JSS

View full text Add to dashboard Cite

This paper proposes a multi-indicator search study based on the social network perspective for borrowing and lending customers. When studying P2P lending and losing customers, according to the division of interpersonal relationship in sociology and the extent to which the client can be debited in reality, it is divided into three kinds of relationship networks: blood relationship network, geo-relationship network and business relationship network, assisting to combine the two-mode subordinate network of the lost customers, and focusing on analyzing its scale, density and other indicators, and hope to find the target lost customers. Key figures in the social network of interpersonal relationships, message communicators, behavioral influencers, etc., to achieve the search for lending customers after their sudden loss. For the first time, this paper applies the social network theory to the search for borrowed and lost customers and proposes the MIL algorithm. It proposes a new idea for the hiddenness of the loan client's loss of connection, and accelerates the governance of the ill-fated phenomenon of loan and loan loss effectiveness. This paper also analyzes the case and proposes relevant suggestions in connection with a loss event in Guangzhou.

show abstract

Section: Introductionmentioning

confidence: 99%