Prediction and Analysis of Air Incidents and Accidents Using Case-Based Reasoning

Abstract

Prediction of upcoming events has very critical role in many disciplines of life. Air accidents and incidents are one of such critical events. There are many existing learning methods in literature. Case-based reasoning (CBR) is a lazy learning technique of artificial intelligence which exploits past experience very efficiently. It works well when precise information is not available and available information is not well-structured. In this paper, we propose to apply CBR for prediction of air accidents and incidents. In the proposed framework, we describe the retrieval strategies, solution algorithms and revision mechanism. We have implemented the proposed idea for the data of air accidents, incidents and crashes. The results show that up to 87% accuracy can be achieved using the proposed framework.