Enhancing pattern recognition in social networking dataset by using bisecting KMean

Abstract

Databases today can range in size more than terabytes. Within these masses of data lies hidden information of strategic importance. So when there are lots of trees, how to find conclusions about the forest? The newest answer is mining of data, which is being used to increase revenues. Data mining is a process that uses a variety of data analysis tools to discover patterns and relationships in data that may be used to make valid predictions. This research uses social networking data set for pattern recognition, because it is one of the emerging application areas in data mining. We used Facebook 100 dataset and applied Bisecting KMeans algorithm on it, so that we would get better clustering outputs. Bisecting KMeans first bisects the data into 2 parts and selects the part with greater number of elements, then apply clustering on it again. This goes on till we have N Number of clusters. We would apply this to our dataset to get desired results. With this we are going to compare Bisecting K Mean algorithm with other data mining algorithm. And finally we are going to find out different pattern from social networking dataset.