Avoiding Malicious Agents in E-Commerce Using Fuzzy Recommendations

This paper presents a complex computing reputation mechanism to be applied in a domain of e-commerce, that handles reputation as a fuzzy set while decision making is inspired in a cognitive human-like approach. This reputation algorithm has been proved in several scenarios where the goal is to avoid (in other words, filter out) the disturbing influence of wrong recommendations from agents with malicious intentions, such as a collusion between a merchant and a recommender. Results show how the evolution of reputation computed with our fuzzy reputation algorithm is compared with other previous reputation models.     

New membership function for poverty measure

Fuzziness in a fuzzy set is determined by its membership function (m.f) which translates the reality of a problem. Accordingly, the shapes of membership functions (m.fs) are important for a particular problem such as poverty since they effect on a fuzzy inference system. Some authors have used to visualize the behaviour of poverty, different shapes like triangular, trapezoidal. In this paper, a specific (m.f), named modified logistic membership function better illustrating the complicated reality, is proposed to measure poverty. The modified logistic membership function is first formulated for several states of poverty and its flexibility in taking up vagueness in poverty is established by an analytical approach using aggregate operators in order to infer a logical conclusion measuring poverty. An application based on individual well-being data from Tunisian households in 2010 is presented to illustrate use of proposed concepts.