Development of graph theoretic models for business promotion on social networks

Abstract

A social network is a social structure made up of a set of actors (such as individuals or organizations) and the dyadic ties between these actors. It is becoming increasingly an important part of marketing platform of any business. Its methodical analysis has gained significant amount of interest for researching not only at the field of modern sociology but also at communication studies, information science, network science etc. Different marketing managers adopt different strategy to market their product on social networks. They face a lot of problems while marketing their products on social networks. On the other hand, a social network is nothing but a large graph, and hence a graph G = (V;E) can be represented as a social network where each vertex v 2 V represents a user and there exists an edge (u; v) 2 E if the users corresponding to vertices u and v are acquaintance. Due to the ability of mapping a social network to a graph, it came into our mind that some of the social network marketing problems could be mapped to graph theoretic problems. In this thesis we list out social network marketing problems those are mappable to graph theoretic problems. We dig into some of the problems to formulate them in details and then we map those problems to graph theoretic problems. For executing a direct marketing campaign which is based on social network data, we examine different cases to cover the whole network with minimum round of campaign. We develop a graph theoretic model for selecting seeds at a viral marketing campaign on a social network where the existance of business rivals is considered. We devise an algorithm to stop propagation of harmful information on a social network. We also identify graph classes based on suitability of campaigning.