ABSTRACT
The development of the information society has been characterized by the proliferation of collaborative highly socialized networks (Arvidsson forthcoming). These collaborative networks have transformed traditional marketing (Cova et al. 2007; Kozinets et al. 2010). Traditionally, purchasing practices have been conceived of as a private act taking place in isolation. To this end, marketing has studied the consumer’s preferences in relation to historical and demographic individual consumption patterns. Nowadays, social and emotional structures in which individuals are embedded are increasingly thought to influence consumers’ attitudes and behavior. The implementation and use of ranking and review systems are perhaps the initial best-known
embodiment of the new perspective. Amazon.com is the most illustrative example. It is the world’s largest online retailer that has based its success on the trust generated by its customer product reviews. The diffusion of social media sites has further enhanced this process. More and more people
gain knowledge about a brand or a product by searching how it is valued and perceived in their social network. Once people have created their own opinion, they share their views and feelings with others and “the subjective truth turns into a collective truth” (Pekka 2010, p. 46). This, in particular, applies to social media platforms, such as Facebook and Twitter that are specifically designed for sharing emotions, feelings and opinions among the users. Consequently, it has become increasingly important for marketers to acquire the ability to
monitor and predict consumer social and emotional behavior in online communities. This has given rise to the emerging marketing discipline of Social Media Analytics, which uses a new set of data mining techniques in order to capture online social and behavioral dynamics. Data mining is the process of automated discovery of hidden patterns and relationships in large datasets. Algorithms, such as real-time trend detection, link prediction and opinion mining, have been deliberately developed to systematically monitor consumer behavior in online social contexts (Pang and Lee 2008).
