| With the rapid development of China’s economy,the increasing demand for electricity,the stealing electricity situation is getting worse and worse.The endless means of stealing has brought great difficulty to the anti-tampering work.The traditional anti-tampering system means have been focusing on manual investigation,It is imperative to rely on software to improve the efficiency of anti-stealing.As a depth learning algorithm,DBN has a great advantage in discovering the deep attribute of the data,it also can find the theft data behind users data through the data analysis.We apply the method of deep learning to the problem of anti-tampering based on the existing situation of stealing and anti-stealing,an anti-tampering model based on DBN algorithm is proposed,and the feasibility of DBN algorithm in anti-tampering problem is verified by experiments.This paper also compares the DBN algorithm and the BP Neural Network algorithm,Decision Tree algorithm,Logical Regression algorithm,Naive Bayesian algorithm based on the large data analysis.And draw the solution that the DBN algorithm is more advantageous in the anti-tampering problem.Finally,this paper design and display by GIS of anti-stealing system based on DBN algorithm and large data platform,and uses the "large data analysis and exploration tools" to build modules,uses GIS to display.This paper assisted in the management of stealing problems with the use of the software,which reduced staff workload and improved the efficiency. |