Service Level Comparison for Online Shopping using Data Mining

The term knowledge discovery in databases (KDD) is the analysis step of data mining. The data mining goal is to extract the knowledge and patterns from large data sets, not the data extraction itself. Big-Data Computing is a critical challenge for the ICT industry. Engineers and researchers are dealing with the cloud computing paradigm of petabyte data sets. Thus the demand for building a service stack to distribute, manage and process massive data sets has risen drastically. We investigate the problem for a single source node to broadcast the big chunk of data sets to a set of nodes to minimize the maximum completion time. These nodes may locate in the same datacenter or across geo-distributed data centers. The Big-data broadcasting problem is modeled into a LockStep Broadcast Tree (LSBT) problem. And the main idea of the LSBT is defining a basic unit of upload bandwidth, r, a node with capacity c broadcasts data to a set of [c=r] children at the rate r. Note that r is a parameter to be optimized as part of the LSBT problem. The broadcast data are further divided into m chunks. In a pipeline manner, these m chunks can then be broadcast down the LSBT. In a homogeneous network environment in which each node has the same upload capacity c, the optimal uplink rate r, of LSBT is either c=2 or 3, whichever gives the smaller maximum completion time. For heterogeneous environments, an O(nlog2n) algorithm is presented to select an optimal uplink rate r, and to construct an optimal LSBT. With lower computational complexity and low maximum completion time, the numerical results shows better performance.The methodology includes Various Web applications Building and Broadcasting followed by the Gateway Application and Batch Processing over the TSV Data after which the Web Crawling for Resources and MapReduce process takes place and finally Picking Products from Recommendations and Purchasing it.