Open AccessIdentification of key nodes in abnormal fund trading network based on improved pagerank algorithm
Author(s) -
Jianying Xiong,
Xiao-Yong Wen
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1774/1/012001
Subject(s) - money laundering , database transaction , computer science , key (lock) , identification (biology) , node (physics) , pagerank , algorithm , value (mathematics) , data mining , computer security , finance , business , information retrieval , database , machine learning , botany , structural engineering , engineering , biology
The traditional network theory is unsuitable for money laundering path analysis because of the complexity of financial transaction. The abnormal accounts are identified only by effectively mining the characteristics of fund flow. The importance of transaction node in the whole illegal capital flow is equivalent to that of the connection among these nodes. Through combining the page rank algorithm with the characteristics of fund flow, an improved weighted and iterative initial value mechanism is designed to calculate the transaction heat value of the account, thus the abnormal capital transaction account is screened out. When the method is applied to a money laundering case, the results show the improved page rank algorithm can effectively screen out the key account nodes in the money laundering network and provide clues for the monitoring of money laundering crimes.