关键词: 生产成本, 钢铁企业, 数据挖掘

Abstract: In order to analyze the factors affecting iron & steel production cost and realize decision-support, the effect of various parameters on the production cost was studied, and production cost data mart was constructed. Team and working procedure were defined as classified attributes, and cost difference of each furnace of steel was considered as mining target. By ID3 algorithms, the information gains from each attribute were computed, and the attribute with the highest information gain was chosen as the test attribute. Attributes were chosen repeatedly in this way until a complete decision tree was generated. The decision tree was used to acquire the cost management rule, which implemented the dynamic cost analysis and achieved the remarkable effects.

Key words: production cost, iron & steel enterprise, data mining, decision tree