AI for Small-Scale Farmers: A Speech Dialog Application for Pesticide Safe Use

Xiaofan Liu of the City University of Hong Kong Shenzhen Research Institute in China will harness AI technologies, including a Large Language Model (LLM) and knowledge graphs, together with a smartphone application to help small-scale farmers with low literacy, use pesticides more effectively and safely. Pesticide use has dramatically increased in the last decade, and its misuse, such as incorrect selection, mixing, and application, has substantial environmental and health impacts. They will build a smartphone application with a back-end system comprising a large database of pesticide manuals, a knowledge graph extracted by the LLM from the manuals, and a natural language query mechanism. Users will be able to voice queries directly into the application, upload photos of pesticide labels or scan barcodes and then receive guidance on their use. Their application will be evaluated with farmers in the field for task completion success rates, ease of use and reliability of query results.