Decomposed and weighted characteristic analysis for the quantitative estimation of mineral resources

Characteristic analysis has been employed as a means of selecting favorable targets for mineral exploration. This paper describes a new version of characteristic analysis that is designed to estimate mineral resources as well as delineating exploration targets. The new version, referred to as decomposed and weighted characteristic analysis, employs a weighting scheme for both samples and variables involved in the model to extract the useful information on tonnages and the order of importance of variables. To construct a decomposed model, an optimum cutting technique for the ordered quantity is developed. The model is demonstrated on a case study of pegmatitic Nb-Ta deposits in China.