Along with the maturing of database technique and the spread of data application as well as the establishment of large-scale database and data warehouse, people start to face the challenge of"sufficient data, but limited information". Data mining technique is needed to find handy patterns from massive data.Data mining is to locate covered, unknown knowledge and pattern potentially useful for strategic decision making from massive data. Those patterns contain the specific relationship among a set of objects in the database, providing useful information and the basis for operation and marketing plan, and financial forecasting. Data mining is an intercrossed subject, involving many fields such as machine learning, model reorganization, induction and deduction, statistics, database and high performance calculation.In accordance with the universality and particularity of coal industry, this paper proposes and establishes a user-based data mining model and applies it to the coal industry.This paper focuses on the algorithms used in the data mining, with the advantages of high accuracy rate and great noise data counteracting ability. Neutral Network Algorithms, among which the BP Neural Network is the most widely used, are commonly used algorithms in data mining technique. Problems such as high noise data can be solved by using the user analysis model based on data mining, which is built with the application of BP Neutral Network Algorithm.However, BP Neutral Network Algorithm has some disadvantages such as low efficiency, slow convergence and being apt to get trapped in local minimum. To solve these problems we propose the genetic algorithms to assist and optimize BP Neural Network, which results in the genetic BP Neural Network. This algorithm avoids the problems mentioned above and solves the specific problems in coal... |