Paper Details

Inflated reputation fraud is a serious problem in online auction. Recent work suggested that neighbor diversity is an effective feature for discerning fraudsters from normal users. However, there exist many different methods to quantify diversity in the literature. This raises the problem of finding the most suitable method to calculate neighbor diversity for detecting fraudsters. We collect four different methods to quantify diversity, and apply them to calculate neighbor diversity. We then use these various neighbor diversities for fraudster detection. Experimental results on a real-world dataset demonstrate that, although these diversities were calculated differently, their performances on fraudster detection are similar. This finding reflects the robustness of neighbor diversity, regardless of how the diversity is calculated.

Laksamee Khomnotai, Jun-Lin Lin, “Detecting Fraudsters in Online Auction Using Variations of Neighbor Diversity”, International Journal of Advance Engineering and Research Development (IJAERD), Vol. 13, Issue 7, pp. 1-10, August 2026.