尹亚晶



摘 要: 为了提高体育评价决策的性能,通过体育关联数据挖掘为系统提供数据支持,进行体育评价决策支持系统优化设计,构建体育评价决策支持系统的总体结构模型,进行系统的功能模块化构架和技术指标分析,在嵌入式Linux的内核结构中进行软件开发,采用VIX总线技术进行体育数据采集。设计基于互信息特征提取的数据挖掘技术。在程序加载模块中进行数据挖掘代码加载,采用Qt/Embedded 4.6创建体育评价决策支持系统形用户接口,实现可视化控制,实现系统的软件开发设计。进行仿真实验分析,结果表明,采用该系统进行体育评价决策信息挖掘的准确度较高,系统的可靠性较好。
关键词: 数据挖掘; 评价决策系统; 系统开发; 互信息特征
中图分类号: TN911.1?34; TP391 文献标识码: A 文章编号: 1004?373X(2017)09?0108?04
Abstract: In order to improve the performance of sports evaluation and decision?making, the data mining of the sports association data provides the data support for the mining system. The optimization design of the sports evaluation and decision support system was carried out. The overall structure model of the sports evaluation and decision support system was constructed to analyze the function modularization framework and technical index of the system. The software was developed in the embedded Linux kernel structure, in which the sports data is acquired with VIX bus technology. The data mining technology based on mutual information feature extraction was designed. The code of data mining is loaded in the program loading module. The Qt/Embedded 4.6 is used to create the user interface of the sports evaluation and decision support system to realize the visual control, and system software development and design. The simulation experiment and analysis were carried out for the system. The results show that the system has high accuracy to mine the sports evaluation and decision information, and high reliability.
Keywords: data mining; evaluation and decision system; system development; mutual information characteristic
0 引 言
采用数据挖掘技术和信息加工技术进行体育评价智能决策,通过前期对运动员的体育关联信息数据采集和体育成绩汇总,为运动员的下一步训练提供数据支持。因此,研究体育评价决策支持系统在提高体育运动训练的科学性和合理性方面具有积极意义[1?2]。
当前的体育评价决策方法主要有神经网络训练方法[3]、SOM自组织映射学习方法和支持向量机方法[4?5],结合嵌入式的集成控制设计,实现数据挖掘和体育评价决策,但是上述方法需要大量的先验数据信息作为参照训练输入,在信息缺乏的情况下数据挖掘的准确度不好,且系统的设计存在集成度不高的问题[6]。……