A novel integrated framework based on modular optimization for efficient analytics on twitter big data

Abstract

In big data, one of the drastically growing and emerging research areas is processing. Big data means a very large amount of data, and a range of methodologies such as big data collection, processing, storage, management, and analysis is included. It is very tough to arrange, manipulate, and retrieve information from the big data due to high variations in sources of data, type, size, and dimensionality. On providing an efficient information retrieval algorithm for big data, a various number of earlier research works were focused. But the percentage of accuracy obtained by the earlier approaches is not up to the level. Hence, it is motivated to design and implement a novel integrated framework for information retrieval in big data with high accuracy in this research work. The entire framework is carried out into various stages such as preprocessing feature selection for dimensionality reduction, clustering, and classification for data mining. Data preprocessing using feature selection by modular optimization-based feature selection (MOBFS) algorithm and classification by multi-class support vector machine (SVM) based algorithm is applied. Artificial immune system (AIS) integrates the functionality of fast searching from simulated annealing to increase the efficiency in finding optimal features speedily in this algorithm. Using Association Rule Mining And Deduction Analysis (ARMADA) tool in MATLAB software, the entire research work has experimented on the various datasets and the results are verified. By comparing the obtained results with the existing research results, the performance of the framework is evaluated.