Research Article
An Optimized Ensemble Machine Learning Framework for Multi-class Classification of Date Fruits by Integrating Feature Selection Techniques
@INPROCEEDINGS{10.1007/978-3-031-48888-7_2, author={V. V. R. Maheswara Rao and N. Silpa and Shiva Shankar Reddy and S. Mahaboob Hussain and Sridevi Bonthu and Padma Jyothi Uppalapati}, title={An Optimized Ensemble Machine Learning Framework for Multi-class Classification of Date Fruits by Integrating Feature Selection Techniques}, proceedings={Cognitive Computing and Cyber Physical Systems. 4th EAI International Conference, IC4S 2023, Bhimavaram, Andhra Pradesh, India, August 4-6, 2023, Proceedings, Part I}, proceedings_a={IC4S}, year={2024}, month={1}, keywords={Date Fruit Classification Machine Learning Feature Selection Ensemble Classifier Neural Networks}, doi={10.1007/978-3-031-48888-7_2} }- V. V. R. Maheswara Rao
N. Silpa
Shiva Shankar Reddy
S. Mahaboob Hussain
Sridevi Bonthu
Padma Jyothi Uppalapati
Year: 2024
An Optimized Ensemble Machine Learning Framework for Multi-class Classification of Date Fruits by Integrating Feature Selection Techniques
IC4S
Springer
DOI: 10.1007/978-3-031-48888-7_2
Abstract
Date fruits are widely consumed and valued for their nutritional properties and economic importance. Accurate classification of date fruits is crucial for quality control, sorting, and grading processes in the food industry. However, the classification of date fruits poses several challenges due to variations in their external visible physical attributes and the presence of multiple cultivars. To address the issue of high-dimensional feature space, the authors, in this research study propose an optimized ensemble machine learning framework for multi-class classification of date fruits by integrating feature selection techniques. The objective is to develop a reliable and efficient system that can accurately classify date fruits into different classes based on their physical attributes. Initially, the framework applies various feature selection techniques to identify the most relevant and discriminative features for classification. Next, it employs variants of ensemble machine learning techniques to perform multi-class classification. We utilize popular ensemble methods such as Boosted Trees, Bagged Trees, RUSBoosted Trees and Optimized Ensemble to improve the accuracy and optimization of the classification model. By integrating feature selection techniques, the proposed optimized ensemble classifier effectively handles the challenges of high-dimensional feature space and variations in date fruit characteristics. The results confirm the superiority of the proposed framework, highlighting its potential for practical applications in the food industry.
