Assessing the Number of Criteria in GIS‐Based Multicriteria Evaluation: A Machine Learning Approach

Geographical Analysis, Journal Year: 2025, Volume and Issue: unknown

Published: March 27, 2025

ABSTRACT The analytical hierarchy process (AHP) is a widely used approach and decision rule to derive criteria weights in geographic information system‐based multi‐criteria evaluation (GIS‐MCE). However, one limitation of the AHP method that it constrains number can be meaningfully weighted typically seven nine criteria. Recently, machine learning (ML) techniques have emerged as compelling alternative for deriving weights. This research aims assess capabilities ML‐MCE handling larger specifically applied case study urban suitability analysis. random forest (RF) ML technique evaluate ability MCE handle up 27 Geospatial data from Metro Vancouver Region, Canada, are used, with subdivided into 11 groups starting most basic incrementally adding two new per group. results indicate RF‐ML manage compared traditional approach, 15 providing meaningful upper threshold, demonstrating its potential accommodate wider range stakeholder preferences complex analysis contexts.

Language: Английский

GIS-Based Agricultural Land Use Favorability Assessment in the Context of Climate Change: A Case Study of the Apuseni Mountains

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(18), P. 8348 - 8348

Published: Sept. 17, 2024

With an emphasis on the effects of climate change, this study offers a thorough GIS-based assessment land use favorability in Apuseni Mountains. The Mountains, region characterized by its biodiversity and complex terrain, are increasingly vulnerable to impacts which threaten both natural ecosystems human activities. territory 11 territorial administrative units was selected for investigation because it shows more anthropogenic influence due migration people mountainous areas following COVID-19 pandemic, increased amount pressure area. Factors that describe area, soil characteristics, morphometric characteristics relief were used create classification present classes restrictiveness plots land, using quantitative GIS model determine main crops agricultural uses. current thus initially obtained, taking into account temperature precipitation values SSP1-1.9, SSP1-2.6, SSP2-4.5, SSP3-7.0, SSP5-8.5 scenarios 2020–2099 time frame. results indicate variation statistical different classes, decrease 4.7% high class pastures, estimated 4.4% grassland, case orchards, situation reflects fluctuating variation. There is 6.4% very low according SSP2-4.5 (in reaching average 12.7 °C annual 895 mm), favorability, there increase falling better up 0.7%.

Language: Английский