Holoentropy based Correlative Naive Bayes classifier and MapReduce model for classifying the big data

Big data is the recent imminent technology, which can provide large benefits to the business administration. Owing to such huge volume, it becomes very complicated to ensure effective analysis by the existing techniques. The complications can be related to analyze, capture, sharing, storage, and vis...

Full description

Saved in:
Bibliographic Details
Published in:Evolutionary intelligence 2022, Vol.15 (2), p.1037-1050
Main Authors: Banchhor, Chitrakant, Srinivasu, N.
Format: Article
Language:English
Subjects:
Applications of Mathematics
Artificial Intelligence
Big Data
Bioinformatics
Business administration
Classification
Classifiers
Conditional probability
Control
Correlation
Datasets
Engineering
Mathematical and Computational Engineering
Mechatronics
Robotics
Special Issue
Statistical Physics and Dynamical Systems
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Big data is the recent imminent technology, which can provide large benefits to the business administration. Owing to such huge volume, it becomes very complicated to ensure effective analysis by the existing techniques. The complications can be related to analyze, capture, sharing, storage, and visualization of the data. To tackle these challenges, a novel classification technique using Holoentropy based Correlative Naive Bayes classifier and MapReduce Model (HCNB-MRM) is proposed. The proposed HCNB, which is designed by combining the Holoentropy function with the correlative based Naive Bayes classifier deals with both high-dimensional data sets as well as extensive datasets to improve the benchmark, and classify the data based on dependent assumption. Therefore, the proposed HCNB-MRM is used to make the process simpler and to choose the best features from big dataset. The proposed HCNB with the MapReduce Model maximizes the performance of big data classification using probability index table, and posterior probability of the testing data samples. The performance of the proposed HCNB-MRM is evaluated using three metrics, such as accuracy, sensitivity, and specificity. From the experimental results, it is analyzed that the proposed HCNB-MRM obtains a high classification accuracy of 93.5965% and 94.3369% for the localization dataset, and skin dataset when compared with the existing techniques.
ISSN:1864-5909
1864-5917
DOI:10.1007/s12065-019-00276-9