Machine Learning Based Prediction and Prevention of Malicious Inventory Occupied Orders

2. Literature Review

While online shopping is becoming more popular, financial fraud is becoming an increasingly serious phenomenon. Criminal activities are driven by enormous economic benefits, and they are also protected by the virtual environment. Phua et al. (2010) defined fraud as “the abuse of profit organizations system without necessarily leading to direct legal consequences.” Bhargava et al. (2003) proposed a technique that identifies auction fraud by detecting abnormal profiles and user behaviors and building patterns from exposed fraudsters in order to discover malicious intentions.

The problem of distinguishing between normal and atypical behaviors is not new and several Artificial Intelligence techniques have been employed to address it. The term classification refers to techniques (Debar et al., 1992), (Jackson et al., 1991), (Lunt et al., 1989), (Sebring et al., 1988), (Teng et al., 1990) which derive patterns of normal activities within a specific domain and then filter data into normal or exceptional categories based on the set of known patterns.