数据挖掘技术在排球生化指标分析中的应用  

Application of Data Mining Technology in Volleyball Athletes' Biochemical Indexes Analysis

作　　者：马静[1] 刘功聚[1] 

机构地区：[1]浙江体育职业技术学院科研处,浙江杭州311231

出　　处：《浙江体育科学》2016年第5期96-99,共4页Zhejiang Sport Science

摘　　要：目的:采用数据挖掘技术分析生化指标与运动员技术等级间的关系,探索决策树算法及关联规则在排球生化分析中应用的可行性。方法:应用IBM SPSS Modeler 14.1软件中C5.0决策树及Apriori关联规则算法,对30名浙江省优秀女子排球运动员历年来部分运动生化数据进行分析。结果:生成11层决策树,模型整体预测精度为86%。产生6条关联规则。结论:C5.0决策树与Apriori关联规则算法可用于女子排球运动员技术等级的预测及规律分析。决策树预测分析中,肌酸激酶为关键因素,其次为睾酮及皮质醇。关联规则分析中,睾酮为关键变量。Objective： Analyze the relationship between biochemical indexes and the grade of athletes with data miningtechnology. Explore the practicability of using decision tree and association rule in volleyball athletes？ biochemical indexesanalysis. Methods：We applied C5. 0 and Apriori algorithm for analyzing 30 volleyball female athletes？ biochemicaldata, with the IBM SPSS Modeler 14. 1 software. Results： A decision tree with 11 levels is built, and the model’s prediction accuracy is 86%. 6 association rules are found. Conclusion： C5. 0 and Apriori algorithm can be used forthe prediction and regularity analysis of the volleyball female athletes？ grade. Creatine kinase is the most importantindex for the decision tree, and the testosterone and cortisol are in the next place. Testosterone is the most importantindex for association rule analysis.

关 键 词：数据挖掘 决策树 关联规则 生化指标 

分 类 号：G804.7[文化科学—运动人体科学]