Unveiling Big Data Insights: A Neutrosophic Classification Approach for Enhanced Prediction with Machine Learning

Abstract

The ever-growing volume and complexity of Big Data pose challenges for traditional 
classification tasks. This paper explores the potential of Neutrosophic Sets (NS), a powerful 
framework for handling uncertainty, in building robust classification models for Big Data prediction 
using Machine Learning (ML) techniques. We provide a detailed background on NS and discuss its 
advantages over Fuzzy Sets. We then propose a methodology that integrates NS with relevant ML 
algorithms for classification. We evaluate the performance of our Neutrosophic-based model on a Big 
Data source. The results are analyzed to assess the effectiveness of the Neutrosophic approach for Big 
Data prediction. This research contributes to the advancement of uncertainty management in Big 
Data classification and paves the way for further exploration of Neutrosophic-based ML models for 
various prediction tasks. Results show that the Neutrosophic Neural Networks (NNs) model 
achieved commendable performance across various metrics, with an accuracy of 79.08%, precision of 
74.58%, recall of 77.64%, and an F1-score of 75.63%. These metrics indicate that the Neutrosophic NNs 
model effectively balances the trade-offs between precision and recall, providing a robust 
classification performance in the context of the evaluated dataset

DOI: 10.5281/zenodo.13381926