<b>Beyond absolute space: Modeling disease dispersion and reactive actions from a multi-spatialization perspective</b>
收藏资源简介：
<b>Overview</b>This document provides instructions on how to use the data and code associated with the manuscript titled “<b>Beyond absolute space: Modeling disease dispersion and reactive actions from a </b><b>multi-spatialization perspective</b>”. The following sections will guide you through the setup, data structure, code execution, expected output, and any additional notes necessary for reproducing the results presented in the manuscript.<b>Table of Contents</b>· Requirements· Data files· Code structure· Running the code· Expected Output· Troubleshooting==========================================================RequirementsOperating system· Windows 7 or higher (recommended)· UbuntuSoftware· Python (version 2.7 or higher) or Jupyter NotebookRequired libraries: numpy, pandas, scipy, matplotlib, pgmpyData filesSurvey_data_processed_Anonymized.csvProtectiveAction_Anonymized.csvThese two data files have been pre-processed from the raw survey data to support the Python code for generating Figures 3, 4, 5, and 6. To protect the privacy and confidentiality of human research participants, all personal information has been excluded in the pre-processing.The data files include anonymized individual record IDs, self-reported weekly symptoms (for themselves and others), protective actions taken, and the service places they visited each week (20 types). The data files also include information regarding the daily volume of visits and the presence of infectious visitors at the 20 types of service places.Code structure· <i>/ Firstlayer_ModifyandUpload.ipynb</i>This is the code file for the first layer of the Bayesian network analysis and SHAP analysis.· <i>/</i><i> </i><i>SecondLayerProtectiveAction.ipynb</i>This is the code file for the second layer of the Bayesian network analysis and SHAP analysis.Running the Code· To run the Python code (preferably in Jupyter Notebook), ensure that all dependencies are installed by running: <i>pip install pandas pgmpy</i>. These dependencies are specified at the beginning of the file.Expected Output· Running the provided Python script will generate the specified figures below. Note that the labels and axis text of the figures are adjusted in the manuscript for readability and to ensure consistency with the manuscript.<i>Firstlayer_ModifyandUpload.ipynb</i>· <b>Figure 3 – Generated in the script (Cell 3, line 85).</b><b>· Figure 4 – Generated in the script (Cell 4, line 101).</b><i>SecondLayerProtectiveAction.ipynb</i>· <b>Figure5 – Generated in the script (Cell 3, line 65).</b><b>· Figure 6 – Generated in the script (line 150). Part of the Figure 5 was generated in ArcGIS Pro.</b>Note:· <b>Table 1 is created directly in Microsoft PowerPoint. Refer to Figures & Table.pptx.</b><b>· Figures 1 and 2 is created directly in Microsoft PowerPoint. Refer to Figures & Table.pptx.</b>TroubleshootingIf you encounter issues while running the script, check the following:· Missing Data Files: Ensure all required data files are in the same directory as the script or that the correct file paths are specified.· Library/Package Errors: Ensure that all necessary libraries and packages are installed. Use pip install as needed.
<b>概述</b> 本文档为论文《<b>超越绝对空间：基于多空间化视角的疾病传播与应对行为建模</b>》配套的数据与代码使用说明。 下文将依次指导您完成环境配置、数据结构说明、代码运行、预期输出以及复现论文中所述结果所需的全部补充说明。 <b>目录</b> · 环境依赖 · 数据文件 · 代码结构 · 代码运行指南 · 预期输出 · 故障排查 ========================================================== <b>环境依赖</b> 操作系统 · Windows 7及以上版本（推荐） · Ubuntu 软件 · Python（2.7版本及以上）或Jupyter Notebook 所需依赖库：numpy、pandas、scipy、matplotlib、pgmpy <b>数据文件</b> Survey_data_processed_Anonymized.csv ProtectiveAction_Anonymized.csv 上述两个数据文件均已从原始调研数据中完成预处理，用于支撑Python代码生成图3、图4、图5与图6。为保护人类研究参与者的隐私与机密性，预处理阶段已移除所有个人身份信息。 数据文件包含匿名化的个体记录ID、自我报告的每周个人及他人症状、已采取的防护行为，以及每周到访的20类服务场所信息。此外，数据还涵盖这20类服务场所的每日到访人流量及传染性访客在场情况。 <b>代码结构</b> · <i>/Firstlayer_ModifyandUpload.ipynb</i> 该文件为贝叶斯网络（Bayesian network）分析与SHAP分析的第一层代码文件。 · <i>/SecondLayerProtectiveAction.ipynb</i> 该文件为贝叶斯网络（Bayesian network）分析与SHAP分析的第二层代码文件。 <b>代码运行指南</b> · 运行Python代码（推荐在Jupyter Notebook中运行）前，请先执行以下命令安装所需依赖：<code>pip install pandas pgmpy</code>。代码文件开头已列明全部依赖项。 <b>预期输出</b> 运行提供的Python脚本将生成下述指定图表。请注意，论文中为提升可读性并确保格式统一，已对图表的标签与轴文本进行了调整。 <i>/Firstlayer_ModifyandUpload.ipynb</i> · <b>图3：由脚本第3个代码单元格第85行生成。</b> · <b>图4：由脚本第4个代码单元格第101行生成。</b> <i>/SecondLayerProtectiveAction.ipynb</i> · <b>图5：由脚本第3个代码单元格第65行生成。</b> · <b>图6：由脚本第150行生成。图5的部分内容由ArcGIS Pro生成。</b> 注意事项： · <b>表1可直接在Microsoft PowerPoint中制作，详见Figures & Table.pptx。</b> · <b>图1与图2可直接在Microsoft PowerPoint中制作，详见Figures & Table.pptx。</b> <b>故障排查</b> 若您在运行脚本时遇到问题，请检查以下内容： · 数据文件缺失：确保所有所需数据文件与脚本位于同一目录，或已正确指定文件路径。 · 库/包错误：确保已安装所有必要的库与包，可根据需要使用pip install命令进行安装。
