Forecasting Insect Pests Through Case-Based Reasoning

Abstract In this paper, FIP, a case—based reasoning system is described. FIP makes its inference based on past cases and learns forecasting knowledge throug h cases to improve the problem-solving capabilities of an expert system. FIP is a reactive forecaster that when gathering new observations can immediately generate a new case to cope with the new situations. As a learner, FIP is capable of learning success and failure during its problem-solving process. The case memory, inference engine, and learning strategy are introduced.