Land Use Suitability Analysis for Agriculture Crops Farming in Kaski District, Nepal: A Conjunction Study of Analytical Hierarchy Process, Random Forest and Support Vector Machine
OODBODHAN · 12.06.2026 · 9 · 48-63
DOI: 10.3126/oodbodhan.v9i1.95646 ↗Kurz gefasst
Fachartikel in der Zeitschrift OODBODHAN, online erschienen am 12.06.2026. Verfasst von Yogendra Karki und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
A mix of machine learning algorithms and expert-based Analytical Hierarchy Process was applied to estimate land suitability for agricultural crop production in Kaski District, Nepal. For land parcels such as Kaski, in which cases urbanization and land fragmentation exert further pressures on agricultural land, a scientific analysis of land appraisal is needful to ensure support for sustainable agricultural practices in a country like Nepal, which has a strong agricultural-centric economy and differing physiographic conditions. Three methods were adopted for suitability mapping: Support Vector Machine classification with robust classification performance, Random Forest classification in which intricate relationships among environment-driven factors are easily managed, and Analytical Hierarchy Process, which favors expert knowledge using pairwise comparison matrices. By adopting GIS-based weighted analysis, all eighteen criteria such as weather conditions, soil qualities, topographic factors, and infrastructure accessibility were analyzed together through land appraisal analysis. Both confusion matrices and classification reports were adopted to ensure correct functioning of this model, and analysis revealed superiority of machine learning algorithms over AHP alone as supporting analysis methods. Climatic conditions and soil productivity emerged as important factors, with Southern and Southeastern sections being most favorable, which covered about 9.68% of this land parcel. Difficult topographic and environmental conditions made north sections most prone to low to moderate adaptability conditions only. As far as land planning decisions and agricultural extension support systems in Kaski District of Nepal are concerned, this combined analysis helps to provide support systems with spatiality, and this has very useful practical applications in mountainous sections interested in prioritizing agricultural land allocation through expert knowledge and latest computer technology-based interventions.
Zitieren
APA
Karki, Y., Katuwal, N. B., & Gautam, S. (2026). Land Use Suitability Analysis for Agriculture Crops Farming in Kaski District, Nepal: A Conjunction Study of Analytical Hierarchy Process, Random Forest and Support Vector Machine. OODBODHAN, 9, 48-63. https://doi.org/10.3126/oodbodhan.v9i1.95646
Vancouver
Karki Y, Katuwal NB, Gautam S. Land Use Suitability Analysis for Agriculture Crops Farming in Kaski District, Nepal: A Conjunction Study of Analytical Hierarchy Process, Random Forest and Support Vector Machine. OODBODHAN. 2026;9:48-63. doi:10.3126/oodbodhan.v9i1.95646
BibTeX
@article{Karki2026,
author = {Karki, Yogendra and Katuwal, Netra Bahadur and Gautam, Saurav},
title = {Land Use Suitability Analysis for Agriculture Crops Farming in Kaski District, Nepal: A Conjunction Study of Analytical Hierarchy Process, Random Forest and Support Vector Machine},
journal = {OODBODHAN},
year = {2026},
volume = {9},
pages = {48--63},
doi = {10.3126/oodbodhan.v9i1.95646}
}
EUVIMED