Based On Rough Set Theory Of Data Mining Methods In Slope Safety Evaluation

As a new product of intelligent information processing technique, the Rough Set theory is a new method of analyzing, ratiocinating, learning and discovering rule for incomplete data problem which developed by Polish scientist-professor Z. Pawlak. Based on the technique of Rough Set theory, data mining can educe the condition and result(decision- making) properties of knowledge expression system, and find all minimum decision algorithms according to the knowledge.The essence of data mining is based on the function of extracting undiscovered but useful knowledge from the mass amount of, incomplete, unorderly, fuzzy and stochastic data. It's well-known, the expert system which depends upon manual basis has a bottleneck problem when obtaining knowledge of experience, further more, it is also difficult to obtain useful knowledge for complex project of slope; Although artificial neural network system has more network model, the majority has its accommodative limitation. So, there must be several complementary methods, then the anticipative goal can be gained.In allusion to catastrophic failure of artificial and natural slopes, this paper combines with abundant slope failure's diachronic data, employs the technique of data mining and analytical procedure. So the inherent law in diachronic slope failures can be excavated, and those valuable mass data of historical slope failure can be availably used in solving those problems of present projects. In this way, it can solve the difficulties on assembling those irregular data, making those isolating data associated and transforming those useless data to useful information, and so on. Also,it makes the safety evaluation result more accordant to engineering practice by combination of historical and present data.Finally, this paper builds risk analysis model which based on the result of data mining. Further more, by qualitative and quantitative risk analysis, this paper actualizes the comprehensive study of slope risk's natural characteristic, disaster consequence, types of uncertain factors and their effect on decision-making process, and accurately gains the reasonable safety assessment result, also gets the goal of realizing slope's safety evaluation.