基于ADTree改进算法的轮胎大数据质量分析  

Quality Data Analysis of Tyre Industry Based on Optimized ADTree Algorithm

作　　者：许晓彬 李敏波[1,2] XU Xiao-Bin1, LI Min-Bo1,2(1.Software School, Fudan University, Shanghai 200433, China;2.Shanghai Key Laboratory of Data Science, Fudan University, Shanghai 200433, China)

机构地区：[1]复旦大学软件学院,上海200433 [2]复旦大学上海市数据科学重点实验室,上海200433

出　　处：《计算机系统应用》2018年第11期27-34,共8页Computer Systems & Applications

基　　金：国家自然科学基金(61671157);上海科技创新行动计划(18511107800)~~

摘　　要：工业企业在生产制造过程中积累了大量的生产数据.海量的工业数据蕴含了价值巨大的信息,通过分析、挖掘这些工业数据能够提升企业数字化管理与质量数据分析能力.本文以轮胎行业制造大数据的应用为背景,分析了轮胎行业制造大数据的质量分析需求与数据特征,将轮胎生产各个环节的多源异构数据有效整合,经过数据预处理流程,构建了结构化的生产制造与质量检测关联分析数据集.针对传统ADTree算法性能较低的问题,本文使用优化后的自底向上的归纳方法进行了改进,充分利用已知数据,减少了建树时分裂测试评估的计算量.实验证明,改进后的ADTree算法更适用于大数据量的数据挖掘. ADTree的挖掘结果经过整理,可以找出影响轮胎质量的重要因素.Industrial enterprises have accumulated a large amount of production data. Massive industrial data contain valuable information. By analyzing and mining these industrial data, enterprises can enhance the ability of digital management and quality data analysis. This paper analyzes the demand and data characteristics of big data in tyre industry. First, the multi-source and heterogeneous data in every link of tyre production is integrated. After analyzing the data pre-processing process, we build the analysis data set of structured manufacturing and quality inspection. According to the low performance of the traditional ADTree algorithm, this study uses bottom induction method to make full use of the known data and reduce the amount of calculation. The experiment shows that the improved algorithm is more suitable for a large amount of data. After sorting out the results of ADTree, the important factors that affect the quality of the tires can be found.

关 键 词：工业大数据 质量分析 ADTree 数据挖掘 决策树 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]