CAJ | 학술논문

为了能够利用常规测试数据快速准确地确定人体体质状况，引入了纯区间的概念，并提出了一种趋势选择方法，在此基础上提出基于趋势选择的SPRINT算法（TESTSPRI—NT）。利用该算法可以根据身高、体重、握力等基本测试数据快速建立人体体质状况决策树，为快速确定人体体质健康状况提供依据。实验和理论分析结果均表明，提出的TESTSPRINT算法可以有效为分析人体体质健康状况提供决策依据，具有较高的准确性，并且时间和空间开销较原疆）RINT算法小。
위료능구이용상규측시수거쾌속준학지학정인체체질상황，인입료순구간적개념，병제출료일충추세선택방법，재차기출상제출기우추세선택적SPRINT산법（TESTSPRI—NT）。이용해산법가이근거신고、체중、악력등기본측시수거쾌속건립인체체질상황결책수，위쾌속학정인체체질건강상황제공의거。실험화이론분석결과균표명，제출적TESTSPRINT산법가이유효위분석인체체질건강상황제공결책의거，구유교고적준학성，병차시간화공간개소교원강）RINT산법소。
In order to determine the human physical condition using the conventional tested data quickly and accurately, this paper proposed a trend selection based scalable parallelizable induc- tion of decision trees algorithm （TESTSPRINT）, based on the concept of pure interval and trend selection method. We can create a human physical condition decision tree quickly based on the basic test data such as height, weight and grip strength, according to the decision tree we can determine human physical health status quickly. Theoretic analysis and experimental dem- onstrations show that the algorithms proposed in this paper outperforms existing algorithms in time and space complexity, and it was proved fruitful applications in the decision human physi- cal health status with high accuracy.