面向数据挖掘的图像检索教学演示系统设计
2016-01-27王华秋
王华秋
摘 要: 以往的数据挖掘教学没有把理论知识和工程项目联系起来,抽象的算法理论知识消减了学生的学习兴趣,导致大部分学生无法运用数据挖掘工具解决实际决策问题。文章以数据挖掘中的邻近支持向量机算法为例,开发了图像检索教学演示系统,分析了系统各模块的功能、工作流程和技术流程,介绍了系统的数据仓库管理模块、数据预处理模块和核心算法模块,给出了应用实例。该教学演示系统有助于学生对所学算法知识的深入理解、记忆和巩固。
关键词: 数据挖掘; 邻近支持向量机; 图像检索; 教学演示系统
中图分类号:G642 文献标志码:A 文章编号:1006-8228(2016)01-94-04
Design of image retrieval teaching demonstration system based on data mining
Wang Huaqiu
(Computer Science and Engineering College, Chongqing University of Technology, Chongqing 400054, China)
Abstract: The theoretical knowledge of data mining is not linked to engineering projects in the previous teaching process. The abstract theoretical knowledge of algorithms has weakened the students learning interest, resulting in most of students cannot use data mining tools to solve practical problems. In this paper, the proximal support vector machine algorithm in data mining is used as an example and an image retrieval teaching demonstration system is developed. The system's whole structure, working flow and technology flow are analyzed, the system's data warehouse management module, data preprocessing module and core algorithm are introduced, and application examples are given. The teaching demonstration system will be helpful for students to further understand, memory and consolidate the algorithm knowledge learned.
Key words: data mining; proximal support vector machine; image retrieval; teaching demonstration system
0 引言
如今快速增长的大数据已经远远超出人们的理解能力,因此大数据挖掘技术受到了广泛关注。只有挖掘和运用海量数据,获得有价值的知识和信息,才能帮助人们制定正确的决策[1-2]。很多高校为工程类研究生开设了数据挖掘这门课程,研究如何将数据挖掘技术运用于企业管理决策的实际中去。
根据理工科高校培养应用型人才的目标,该课程不仅要求学生掌握数据挖掘算法知识,还要以“工程化”理念培养学生的工程技能[3-5]。而现实教学却让数据挖掘课变成了一门算法分析课或学术讨论课,教师教得乏力,学生学得乏味,没能达到人才培养目标。
在研究生数据挖掘课程的教学中,不能单纯地讲解数据挖掘算法证明,也不能简单地计算和人为构造数据挖掘算法的例子、习题。……
