Social Network Analysis using Interest Mining: A Critical Review

Background/Objectives: Social Networking has been entertaining people for sharing their common ideas and proposals which are analyzed through social relations among them. The problem in the field of social network analysis is the absence of adequate computing resources to handle huge amount of data on World Wide Web. Therefore, users are unable to gather needed information correctly and thereby, the aim is to locate right information at the right time and delivering it to distinct group of people. Methods: Present paper gives the insight into the existing deployment of social network analysis and various ranking techniques which have been devised by various researchers for the social networking capabilities over the network. In order to accomplish the aim, virtual environment can be created for social network analysis. This analysis can be performed by various mining methods such as opinion mining, expert mining, etc. and ranking techniques like object average rating, neighbour variance rating, random rating and many more. Although these techniques optimize the information overload problem accordingly, still there is a need for expert identification. Findings: The future enhancement for social network analysis includes collaborative thinking. Social Network Analysis gathers people having similar interest by creating collaboration among users. This collaboration leads to resource sharing in an efficient manner after the creation of virtual environment. Furthermore, the field of social network analysis may take a turn to link analysis and its various algorithms like Page Rank, Weighted Page Rank and Weighted Page Content Rank which will further help in finding the expert and enhances the information effectively. Application/Improvements: The application to social network analysis is to discover the network of innovators in a regional economy, enhancing dark web analysis and spam behaviour detection. The arduous task of expert identification is an upcoming trend that can be implemented through virtual environment.