Social Network Analysis in Telecommunications

CHAPTER 5

Telecommunications Environment

The highly competitive telecommunications market demands that companies establish and maintain a good customer relations process. There are several traditional approaches to address business issues using behavioral segmentation and predictive models. These traditional approaches, such as clustering and predictive models, are well-established unsupervised and supervised learning approaches, respectively. Although these models can be quite effective in terms of predictions or even in terms of customer understanding, they only present an isolated view of customers. They do not consider customers’ relationships or the influences among the customers, especially over time. Social network analysis can be used to enhance knowledge about customers’ influences in an internal community. New propositions to evaluate customers can clearly distinguish aspects of the virtual communities inside the telecommunications networks, allowing companies to deploy more effective action plans to better spread their products and services. The evolutionary perspective should consider network behavior in distinct time frames, allowing a more complex cross-analysis and also the opportunity to establish a correlation between the strength of the relationship and the influence over time for some particular business events, such as bundle diffusion. Remembering that influence happens first when the relationship is stronger can enable companies to prepare a marketing campaign ...