The Research And Implementation Of NMF-Bsed Technology On Mineral Resources Prediction

Mineral prediction, the evaluation of the potential mineral resources, is the product of the combination of geology, mathematics, information technology and computer technology. Based on the summarization of the metallogenic rules, mineral prediction can point out the possible existing mineral resources and deposits which have not been found, and perform evaluations and predictions of their qualities and quantities with the application of geological theory and other possibly involved methods, such as the techniques for geophysical and geochemical exploration, remote sensing and mathematical geology. With the increasing difficulty of the discovery of mineral resources and the development of modern scientific theories and technological methods, GIS technology nowadays is playing a more and more important role in the field of mineral resources prediction. The appropriate use of such technology and the method of data visualization can greatly improve the efficiency and accuracy of mineral forecast.Theory of Non-Negative Matrix Factorization (NMF) is a matrix decomposition method newly proposed in recent years. It adds the non-negative constraints to the original matrix operation in order to ensure non-negative characteristics and result of data matrix decomposition. Moreover, NMF algorithm is simple and easy to implement and it has features such as dimension-lowering and sparse convergence. Such new method can solve many practical problems;therefore, it is suitable to be applied to the field of mineral prediction with non-negative property and large scale of data.The followings are the main work done in the present paper.First, the author introduces the domestic and overseas status of mineral resources prediction, and explains methods related to mineral forecast.Second, it is elaborated that the origin and characteristics of non-negative matrix. Also, several improved algorithms based on the traditional ones are introduced, the appropriate ones of which are selected to be applied to the field of mineral prediction.Finally, VC ++ used as the tool, this paper designs and developed the visual system of mineral prediction with the Geographic Information System MapGis SD second development library. This system is employed to quantitatively predict 1:200,000 scale silver mineral prediction in the east of Inner Mongolia. Main features of the matrix vectors are extracted by using the selected local non-negative matrix factorization method (LNMF). Compared with the results drew from other statistical methods, it is proved that satisfactory result can be concluded.