Data Stream Mining Algorithms in Big Data: A Survey

The infrastructure build in the big data platform is reliable to challenge the commercial and noncommercial IT development communities of data streams in high dimensional data cluster modeling. The APSO ie., Accelerated Particle Swarm Optimization is a technique which commonly known for data's are sourced to accumulate their continuation in the batch model induction algorithms which is not feasible for the real time applications[8]. In this project, a new technique has been introduced ie., supervised machine learning methods for developing dynamic resource allocation which targets a user defined learning method to identify the workload patterns and also feature selection is used to process the loaded data in the searched space to form the subset of the optimal solution in size to interact their demands in computation. The main theme of this project is to feed up the data in a lightweight feature selection and to designed the streaming data by using APSO, which enables the swarm search layered forms related query dependent performance in the process scheduling and data accuracy in the iterative manner. Thus the Big data in APSO are put under the test of new feature selection algorithm for performance evaluation.