Supplementary material for "A novel interpretable artificial intelligence (XAI) approach for revealing the spatiotemporal variation and driving factors of soil salinization in a semi-arid region"

The repository contains the primary input files utilised to simulate the soil salinization risk model. These files encompass soil salinization point locations and environmental variables, which have been meticulously collected from the extant literature. In addition, the complete Python code for the RF model, the three XAI techniques, factor analysis, and analysis of covariance utilised in the paper are provided. These are all summarised in the order in which they are presented in the paper (Phase Ⅰ to Ⅳ).

本仓库包含用于土壤盐渍化风险模型模拟的核心输入文件。这些文件涵盖土壤盐渍化点位坐标与环境变量数据，所有数据均从现有文献中精心搜集整理而来。此外，本仓库还提供了论文中所使用的随机森林（Random Forest, RF）模型、三种可解释人工智能（eXplainable Artificial Intelligence, XAI）技术、因子分析以及协方差分析的完整Python代码。所有内容均按照论文中的呈现顺序（第一阶段至第四阶段）进行整理汇总。