Improving Classification Accuracy in Using Evolutionary Decision Tree Filtering in Big Datasets

Nikhat Khan, Ashish Jain

2023

Abstract

Large datasets frequently have data value errors, which can be a serious issue. The method described in this paper uses a classification filter to locate and eliminate incorrect training instances from large datasets. The suggested approach uses an evolutionary decision tree technique as a filter classifier and aims to increase classification accuracy by pursuing a broad search in the problem space using a genetic algorithm. This method improves classification accuracy compared to conventional decision tree algorithms by using classification accuracy as a fitness function instead of local greedy search on data sets. The proposed approach performed better than earlier methods in terms of decision tree size and classification accuracy after being tested on the UCI repository data set. The results demonstrate the effectiveness of the proposed method in handling datasets with errors in data values.

Paper Citation

in Harvard Style

Khan N. and Jain A. (2023). Improving Classification Accuracy in Using Evolutionary Decision Tree Filtering in Big Datasets. In Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT; ISBN 978-989-758-661-3, SciTePress, pages 189-191. DOI: 10.5220/0012610000003739

in Bibtex Style

@conference{ai4iot23,
author={Nikhat Khan and Ashish Jain},
title={Improving Classification Accuracy in Using Evolutionary Decision Tree Filtering in Big Datasets},
booktitle={Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT},
year={2023},
pages={189-191},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0012610000003739},
isbn={978-989-758-661-3},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Artificial Intelligence for Internet of Things: Accelerating Innovation in Industry and Consumer Electronics - Volume 1: AI4IoT
TI - Improving Classification Accuracy in Using Evolutionary Decision Tree Filtering in Big Datasets
SN - 978-989-758-661-3
AU - Khan N.
AU - Jain A.
PY - 2023
SP - 189
EP - 191
DO - 10.5220/0012610000003739
PB - SciTePress