气候变化领域集成服务门户(CCPortal): Mapping of soil suitability for medicinal plants using machine learning methods

CCPortal

DOI	10.1038/s41598-024-54465-3
	Mapping of soil suitability for medicinal plants using machine learning methods
	Roopashree, S.; Anitha, J.; Challa, Suryateja; Mahesh, T. R.; Venkatesan, Vinoth Kumar; Guluwadi, Suresh
发表日期	2024
ISSN	2045-2322
起始页码	14
结束页码	1
卷号	14 期号:1
英文摘要	Inadequate conservation of medicinal plants can affect their productivity. Traditional assessments and strategies are often time-consuming and linked with errors. Utilizing herbs has been an integral part of the traditional system of medicine for centuries. However, its sustainability and conservation are critical due to climate change, over-harvesting and habitat loss. The study reveals how machine learning algorithms, geographic information systems (GIS) being a powerful tool for mapping and spatial analysis, and soil information can contribute to a swift decision-making approach for actual forethought and intensify the productivity of vulnerable curative plants of specific regions to promote drug discovery. The data analysis based on machine learning and data mining techniques over the soil, medicinal plants and GIS information can predict quick and effective results on a map to nurture the growth of the herbs. The work incorporates the construction of a novel dataset by using the quantum geographic information system tool and recommends the vulnerable herbs by implementing different supervised algorithms such as extra tree classifier (EXTC), random forest, bagging classifier, extreme gradient boosting and k nearest neighbor. Two unique approaches suggested for the user by using EXTC, firstly, for a given subregion type, its suitable soil classes and secondly, for soil type from the user, its respective subregion labels are revealed, finally, potential medicinal herbs and their conservation status are visualised using the choropleth map for classified soil/subregion. The research concludes on EXTC as it showcases outstanding performance for both soil and subregion classifications compared to other models, with an accuracy rate of 99.01% and 98.76%, respectively. The approach focuses on serving as a comprehensive and swift reference for the general public, bioscience researchers, and conservationists interested in conserving medicinal herbs based on soil availability or specific regions through maps.
英文关键词	Choropleth; Decision tree; Extra trees classifier; Geographic information system; Machine learning; Medicinal plants; Supervised learning
语种	英语
WOS研究方向	Science & Technology - Other Topics
WOS类目	Multidisciplinary Sciences
WOS记录号	WOS:001163287600007
来源期刊	SCIENTIFIC REPORTS
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/302675
作者单位	Jain University; Vellore Institute of Technology (VIT); VIT Vellore; Adama Science & Technology University
推荐引用方式 GB/T 7714	Roopashree, S.,Anitha, J.,Challa, Suryateja,et al. Mapping of soil suitability for medicinal plants using machine learning methods[J],2024,14(1).
APA	Roopashree, S.,Anitha, J.,Challa, Suryateja,Mahesh, T. R.,Venkatesan, Vinoth Kumar,&Guluwadi, Suresh.(2024).Mapping of soil suitability for medicinal plants using machine learning methods.SCIENTIFIC REPORTS,14(1).
MLA	Roopashree, S.,et al."Mapping of soil suitability for medicinal plants using machine learning methods".SCIENTIFIC REPORTS 14.1(2024).