Applying case-based reasoning for mineral resources prediction

Abstract

Case-Based Reasoning (CBR), a well known Artificial Intelligence (AI) technique, which consists of retrieving, reusing, revising, and retaining cases, has already proven its effectiveness in numerous industries. In this research, we try to adopt CBR technique in mineral resources prediction. A model for mineral resources prediction is proposed in this paper, which can support the processes of case-based reasoning in mineral resources prediction such as case representation, indexing, retrieving and case revising. It mainly includes Feature tree and FSM algorithm and it is different from traditional model. At last, an experiment of iron resources prediction is performed in Eastern Kunlun Mountains, China. The results indicated that the model proposed in this paper is suitable for regional metallogenic prediction. © 2012 Springer-Verlag Berlin Heidelberg.