A high-precision catalogue of landslide events in China based on news text mining with large language model

Key Points

• The research aims to create a precise catalogue detailing landslide events in China using news data.
• Developed a pipeline for extracting information from unstructured news reports
• Implemented text mining techniques using a large language model
• Conducted spatial-temporal deduplication for data accuracy
• Executed high-precision geocoding for event localization
• Generated a comprehensive list of landslide events with high accuracy
• Automated the extraction process significantly improved event detection speed
• Demonstrated effectiveness of the large language model in processing text data