Abstrakt

Fraud Detection Using A New Multilayered Detection System

Kaminee Gurav, Manisha Gurabe, Priyanka Suryawanshi, Prof.Sinu Mathew

Identity theft is a form of stealing someone's identity in which someone pretends to be someone else, usually as a method to gain access to resources or obtain benefits in that person's name. Identity crime is prevalent, and costly; and credit application fraud is a specific case of identity crime or identity theft. The existing non-data mining detection systems that uses business rules and scorecards, and known fraud matching have limitations. To overcome these limitations and combat identity crime in real-time, we propose a new multi-layered detection system consisting of communal detection (CD) and spike detection (SD) layers that are resilient. Resilience is the longterm capacity of a system to deal with change and continue to develop communal detection (CD) finds real social relationships to decrease the suspicion score, and is tamper-resistant to the synthetic social relationships. It is the whitelist oriented approach on a fixed set of attributes [1]. The CD algorithm matches all links against the whitelist to find communal relationships and reduce their link score. CD can detect more types of attacks; better account for changing legal behavior and spike detection (SD) complements CD.