World Heritage documents reveal persistent gaps between climate awareness and local action

This toolkit is designed for processing and analyzing documents from the UNESCO World Heritage Centre. It supports batch scraping, downloading, verification, and conversion of PDF files into plain text, as well as the extraction of climate-related keywords. The cleaned text outputs are suitable for downstream tasks such as text mining and climate-awareness analysis. The documents are sourced from publicly available official reports. The analysis section includes a GLM model implemented in R, along with evaluation tools such as correlation heatmaps, ICC agreement analysis, and MCC-based binary classification assessment. All tools support customizable labels and visualization of confidence intervals. Some components require non-standard Python libraries such as pdfminer.six and pingouin.

本工具包专为处理与分析联合国教科文组织世界遗产中心（UNESCO World Heritage Centre）的文档而打造。其支持批量爬取、下载、校验，并可将PDF文件转换为纯文本，同时支持气候相关关键词的提取。经清洗后的文本输出可适配文本挖掘、气候感知分析等下游任务。所用文档均源自公开可获取的官方报告。分析模块包含基于R语言实现的广义线性模型（Generalized Linear Model，GLM），以及相关性热图、组内相关系数（Intraclass Correlation Coefficient，ICC）一致性分析、基于马修斯相关系数（Matthews Correlation Coefficient，MCC）的二分类评估等评估工具。所有工具均支持自定义标签与置信区间可视化。部分组件需依赖pdfminer.six、pingouin等非标准Python库。