基于支持向量机的软件工程实验智能评价分析方法研究  

Research on SVM⁃based intelligent evaluation analysis method for software engineering experiment

作　　者：于万国[1,2] 隋丽娜[2] YU Wanguo;SUI Lina(Hebei University,Chengde 067000,China;Hebei Normal University for Nationalities,Chengde 067000,China)

机构地区：[1]河北大学,河北承德067000 [2]河北民族师范学院,河北承德067000

出　　处：《现代电子技术》2021年第22期183-186,共4页Modern Electronics Technique

基　　金：2017年度河北省科技厅项目(17K50320D);2019年度国家民委高等教育教学改革研究项目(19064);2019河北省高等教育学会高等教育科学研究“十三五”规划课题项目(GJXH2019⁃122)。

摘　　要：针对软件工程实验课程采用传统方法评价主观性强,而现有智能评价方法存在准确度低、实用性差等问题,文中在支持向量机等多种智能算法理论的基础上,提出一种具有多个维度的智能评价分析方法。在当前软件工程实验课程组织形式的基础上,通过引入支持向量机和数理统计学理论,从时间安排、任务分布、实验工具、指导方式和教学实施等多个方面,提出了软件工程实验课程的多维度综合评价方法。数据仿真测试结果表明,与基于K⁃means算法的实验评价方法相比,文中所提出的评价方法具有更高的评价准确度和实用性。For the software engineering experimental courses,the traditional methods are highly subjective,while the existing intelligent evaluation methods have problems such as low accuracy and poor practicability.An intelligent evaluation and analysis method with multiple dimensions are proposed based on various intelligent algorithm theories such as support vector machine.On the basis of the current organization form of software engineering experimental courses,by introducing the support vector machines and mathematical statistical theory,the multi⁃dimensional comprehensive evaluation method for the software engineering experimental courses are proposed in terms of time arrangement,task distribution,experimental tools,guidance methods and teaching implementation.The data simulation test results show that,in comparison with the experimental evaluation method based on the K⁃means algorithm,the evaluation method proposed in the article has higher evaluation accuracy and practicability.

关 键 词：软件工程 实验课程 智能评价 支持向量机 多维度评价 仿真测试 

分 类 号：TN915-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]