A Comparative Study of Optimization Algorithms for Feature Selection on ML-based Classification of Agricultural Data
CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS(2024)
摘要
In today’s world, agricultural production and operation activities generate a lot of data. As a result, computer-aided agriculture applications have become a hot topic in the study, with various machine learning (ML) algorithms being used to classify agricultural data. This paper presents a comparative study consisting of a combination of ML algorithms with meta-heuristic algorithms for feature selection to improve the classification capability of ML algorithms by finding the features that significantly impact accuracy. We have used six different meta-heuristic algorithms for feature selection. Experiments are conducted on four different agricultural datasets with five classification models. To understand the effect of proposed models, the selected features are fed into the ML algorithms. The results prove that combining ML and meta-heuristic algorithms achieves higher classification accuracy with fewer features on agricultural datasets.
更多查看译文
关键词
Agricultural data,Machine learning (ML),Classification,Meta-heuristic optimization,Feature selection
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要