Knowledge Discovery in Social Networking for Future Filtering Purposes

Social media has become very popular and is able to generate oceans of data which is a goldmine to prospective researches in the real world. Discovering hidden patterns from such data can provide business intelligence. Mining such social media data and learning collective behavior can have very useful utilities in the real world. Many real world applications like advertising, marketing, sales do have their focus on social networking. Scalable learning of collective behavior is very important research area where as the social media data is very vast. Sustainable solution should promote scalability. Object heterogeneity is one of the challenging aspects that can be incorporated in mining tasks so as to gain actionable knowledge. Recently Tang et al. proposed a solution for this using edge-centric approach. In this paper we are influenced by that work and propose a scheme that can solve the problem of object heterogeneity based on the multi-mode network. With this solution the prediction capabilities of the proposed application will be more. We build an application to demonstrate the proof of concept. The application can predict collective behavior with scalable feature and will be useful in real world applications. Index Terms – Social networking, data mining, social dimensions, collective behavior, clustering _____________________________________________________________________________________________