In this century big data manipulation is a challenging task in the field of web mining because content of web data is massively increasing day by day. Using search engine retrieving efficient, relevant and meaningful information from massive amount of Web Data is quite impossible. Different search engine uses different ranking algorithm to retrieve relevant information easily. A new page ranking algorithm is presented based on synonymous word count using Hadoop MapReduce framework named as Similarity Measurement Technique (SMT). Hadoop MapReduce framework is used to partition Big Data and provides a scalable, economical and easier way to process these data. It stores intermediate result for running iterative jobs in the local disk. In this algorithm, SMT takes a query from user and parse it using Hadoop and calculate rank of web pages. For experimental purpose wiki data file have been used and applied page rank algorithm (PR), improvised page rank algorithm (IPR) and proposed SMT method to calculate page rank of all web pages and compare among these methods. Proposed method provides better scoring accuracy than other approaches and reduces theme drift problem.
High Performance Computation of Big Data: Performance Optimization Approach towards a Parallel Frequent Item Set Mining Algorithm for Transaction Data based on Hadoop MapReduce Framework