Decision making improvement in social marketing strategy through dependent multi-dimensional opinion formation

Abstract

One of the most interesting topics in social network research is opinion formation. In this paper we have introduced a new dependent multi-dimensional opinion formation method. This method models agents with several dependent opinions so that modifying the agent's opinion about one issue can affect its opinion about another issue. Agents share their opinion with agents which have trusted them. A directed graph is used for modeling this trust network. It can be modified based on the opinion sharing between agents. By applying this algorithm in the marketing area, Predicted opinions could be used for creating management reports to facilitate strategy definition and decision making. The results of applying this method on the “Epinion” dataset show that agent's opinions converge to a stable state and a set of clusters have been formed in the community. Prediction of agent's opinion can be used for identifying appropriate products and customers and defining right marketing strategy.