Manufacturing material purchasing analysis method based on decision tree algorithm

• Main Authors: HU CHENGHUA, JIANG AIJIA
• Format: Patent
• Language: Chinese; English
• Published: 03.05.2017
• Subjects: CALCULATING; CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION ORPROCESSING OF GOODS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS; TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINSTCLIMATE CHANGE

The invention provides a manufacturing material purchasing analysis method based on a decision tree algorithm. A decision tree serves as a decision analysis method based on a classification idea. The improved decision tree algorithm is used to analyze and predict problems in purchasing manufacturing materials. A front pruning method and a tree depth limitation method are used to limit segmentation of the decision tree, and a rear pruning method is used to construct an optimal decision tree. The optimal decision tree provides a certain purchasing scheme. The algorithm is prevented from infinite diverging effectively due to limitation of the segmentation of the decision tree by the front pruning method and the tree depth limitation method. An information gain standard deviation serves as a limitation condition of the front pruning method, and the accuracy of the algorithm is improved. The optimal decision tree constructed via the rear pruning method is simple, effective, and easy to realize and understand. The