A case-based reasoning approach to local-scale and short-term weather forecast

As an intelligent computing paradigm Case-Based Reasoning (CBR) techniques have been studied and adopted in weather forecast applications recently. Nonetheless, many subtle issues remain to be resolved for CBR to be a practical problem solver. This paper establishes a CBR system for local-scale and short-term weather forecast at an airport. We develop a strategy of knowledge guidance combined with nearest-similarity and design a heuristic similarity function accordingly. Real data based tests have shown the effectiveness and comparative advantages of the proposed method.