Novel ensemble bagging-logistic regression algorithm for NoSQL database security

doi:10.1007/s10489-025-06358-9

Return

Novel ensemble bagging-logistic regression algorithm for NoSQL database security

2025-02-28

PRE

Anuradha Kanade

Amol D. Vibhute *

Shantanu Kanade

DOI：10.1007/s10489-025-06358-9

Original

Original request for help

Save

Abstract

En 中文

In the present era, the use of the Internet has drastically increased in the sharing of digital information. In this case, the digital information is stored using cloud technology or NoSQL databases. However, there is a significant challenge in protecting and managing the cloud and NoSQL-based data and extracting required information from these sources while maintaining the actual information. The network traffic has also increased significantly, which requires more memory and sufficient systems to manage and monitor the influx of Big Data. Traditional relational databases face issues in managing and securing the cloud-based dynamic data generated from various sources. NoSQL databases have recently been used to store and manage dynamic data effectively. However, there are security and privacy issues with the NoSQL databases, which remain challenging to provide. Consequently, in the present study, we propose a novel algorithm that enhances the security of the NoSQL databases and predicts its success rate. Initially, we implemented the Fernet data masking algorithm to secure the NoSQL database. Then, the secured data is classified and predicted using an innovative proposed method called the Ensemble Bagging Classifier-Logistic Regression (EBC-LR) to validate the accuracy of the secured NoSQL database. The experimental outcomes depict that our proposed algorithm achieves 85 percent accuracy, better than traditional methods in enhancing the security of NoSQL databases. Our proposed algorithm can effectively predict secure standard databases with the highest success rate.

Keywords:

Ensemble Bagging Classifier-Logistic Regression (EBC-LR)

NoSQL database

Data masking

Security

Linear discriminant analysis