A Parallelized Social Network Analysis Using Virtualization for Studentís Academic Improvement
“Big Data is the Next Frontier for Innovation, Competition, and Productivity”. Big data is massive and messy and it’s generated at a very fast rate. These characteristics pose a problem for data storage and processing, but focusing on these factors has resulted in a lot navel-gazing. The data which are generated by social network includes structured (10%) and unstructured (90%), which are huge in volume and are becoming big challenge to process and analyze. Big data technology offers significant contributions over technology development. In addition to this, now a day it is also having higher impact on the academic performance of the students. The social networking sites like LinkedIn, Tweeter, and Face book are having such impact among the younger generation. This paper is concerned with student’s involvement in social media and their interaction with others (student to student and student to tutor) through this network. Through this analysis of their interaction and communication over social network it will be helpful to find out their interest and requirements. This analysis also list out the usage of some web services and their impact over the linguistic and academic behavior of the young learners. In social network analysis the structured data are capable of storing and analyze using traditional analysis techniques. Here the unstructured data are from social media are considered as useless. But it is also needed to analyze and extract meaning full information from them. The unstructured huge volume of data could be processed under special environment with additional techniques. It is not feasible to process under traditional environment. Since the social network could be represented by socio graph, it is will be feasible by partitioning the big socio graph into computable size. Then it is taken as input into virtualized memory to analyze. Here the Mapper() and Reducer() programs are executed under virtual environment in distributed manner. The virtualization of process has been used to reduce the space complexity and time complexity. There are tools created to address the problem of big data. Hadoop is one of the best examples for dealing big data in a distributed environment. It involves breaking the huge socio graph into cliques of similar interest and analyzed to identify the students need for their development.
K.Geetha, Dr.A. Vijaya Kathiravan
To read the full article Download Full Article