Mining Professional's Data from LinkedIn

Social media has become very popular communication tool among internet users in the recent years. A large unstructured data is available for analysis on the social web. The data available on these sites have redundancies as users are free to enter the data according to their knowledge and interest. This data needs to be normalized before doing any analysis due to the presence of various redundancies in it. In this paper, LinkedIn data is extracted by using LinkedIn API and normalized by removing redundancies. Further, data is also normalized according to locations of LinkedIn connections using geo coordinates provided by Microsoft Bing. Then, clustering of this normalized data set is done according to job title, company names and geographic locations using Greedy, Hierarchical and K-Means clustering algorithms and clusters are visualized to have a better insight into them.