An Unsupervised Approach of Knowledge Discovery from Big Data in Social Network

Social network is a common source of big data. It is becoming increasingly difficult to understand what is happening in the network due to the volume. To gain meaningful information or identifying the underlying patterns from social networks, summarization is an useful approach to enhance understanding of the pattern from big data. However, existing clustering and frequent item-set based summarization techniques lack the ability to produce meaningful summary and fails to represent the underlying data pattern. In this paper, the effectiveness co-clustering is explored to create meaningful summary of social network data such as Twitter. Experimental results show that, using co-clustering for creating summary provides significant benefit over the existing techniques.