Applications Of Data Mining Technologies In Metallogenic Prognosis And Economic Evaluation

In a sense,from the early establishment of geology at the beginning,it is essentially an information subject,namely a science of collection,recording,analysis and prediction of the earth geological activities and landscape at each scale.Therefore,the development of the information technology and geology have been closely reflected,mathematical geosciences is the result of the digital information revolution reflects in the geology,is almost a seamless transition to the new stage of geosciences and the natural form of geology.In particular,the arrival of the era of big data further promotes the comprehensive development of the digital geology.The emergence of big data and data mining technology provide a very promising solution.The paper takes Yunnan Gejiu super large Sn-Cu-polymetallic depositas the application object,through the research on the improvement methods of data mining,express the mathematical processes ofthe geological metallogenic process,to provide effective and comprehensive technical support for the metallogenic prediction and economic evaluation.The results of this study can be applied not only to the super large Sn-Cu-polymetallic deposits in Yunnan Province,but also can play an important role of metallogenic prognosis and potentialeconomic evaluation in other regions.This paper uses an improved bidimensional empirical decomposition method to decompose the aeromagnetic survey data,expresses the spatial distribution of deposits with a mixed Gaussian model and decomposes it.By comparing the decomposition results at various sampling data scales and the distribution function for the deposit,the characteristic scale interval containing the measurement information that is closest to the distribution of deposits can be identified.Finally,this method was used to analyse a Yunnan Gejiu tin copper polymetallic deposit,using aeromagnetic sample data to calculate the suitable decomposition scale parameters.This approach provides valuable parameters for metallogenic prediction in other areas with aeromagnetic data.At the same time,it provides an example for the application of other types of exploration information.Secondly,this paper uses the directed graph model of the computer graph theory to model the sampling data,and calculate the segmentation of the directed graph by the improved max flow-min cut algorithm.The proposed energy expressions reflect the continuity of the geological property,and also the smoothness of the segmentation boundary.A major advantage of this method is able to make the geological experts converse the prior knowledge about the partition,such as the favorable ore-forming degree or other variables indicating the metallogenic information a lot,to the seeds of the directed graph and embody in the segmentation result by a very intuitive and simple outline operation,which greatly improves the interactivity and editability of the algorithm.This paper uses geochemical data to identify and divide the metallogenic prospectivity areas and evaluate the economic potential.Finally,the element analysis of multi-scale model is combined with the probabilistic fuzzy logic inference engine,which is simulated by Monte Carlo method is applied to evaluate the metallogenic prospect and the economic potential of the Sn deposit in Gejiu area.