The goal of this thesis is to leverage network data for marketing applications. We concentrate on two topics: First, ways to improve predictive accuracy of classification tasks are analyzed by leveraging information about the social network of a customer. Respective methods could be analyzed in two field studies with a telecommunication provider. Second, centrality measures for viral marketing are investigated. In case data about the customer network is available, centrality measures can be used to spread viral marketing campaigns in a social network. Computational experiments are used to compare different centrality measures for the diffusion of marketing messages. A significant lift could be obtained when using central customers in message diffusion, but we also found significant differences in the various centrality measures depending on the underlying network topology and diffusion process.
«
The goal of this thesis is to leverage network data for marketing applications. We concentrate on two topics: First, ways to improve predictive accuracy of classification tasks are analyzed by leveraging information about the social network of a customer. Respective methods could be analyzed in two field studies with a telecommunication provider. Second, centrality measures for viral marketing are investigated. In case data about the customer network is available, centrality measures can be used...
»