AGENT-BASED MODELING OF REPUTATION FOR SELLERS IN AN EMARKETPLACE

Author (s) : C.Joesphina & K.Ellammal, J.Malathi & K.Gomathi

---

Abstract :
In this paper we present an agent-based mechanism for modeling reputation for sellers in an marketplace based on the ratings for transactions. Most traditional online businesses publish reputation profile for traders that reflect average of the ratings received in previous transactions. This method of modeling a reputation profile is highly susceptible to noisy ratings or manipulation by sellers (for example by using shilling) to artificially inflate their ratings. This paper proposes an adaptive ratings-based reputation model that is robust and is not easily affected by strategic behavior or noisy ratings. The model is based on a trader’s transaction history, witness testimony, and other weighting factors. To validate the proposed model a multi-agent system is built to simulate the interactions among buyers and sellers in an electronic marketplace. The performance of the proposed model is compared to that of other traditional models using a simple scenario where the sellers are selling only one type of product.

Keywords : E-marketplace, Reputation

Conference Date : 30/08/2019

Pages : 324

Paper ID : 19315