• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Proceedings of the ACM special interest group for information technology education conference >A Feasibility Study on using Clustering Algorithms in Programming Education Research
【24h】

A Feasibility Study on using Clustering Algorithms in Programming Education Research

机译：在计划教育研究中使用聚类算法的可行性研究

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Designing an experiment for programming education research, in which collecting and interpreting a large number of qualitative data about programmers is required, needs careful consideration in order to validate the experiment. When it comes to finding a pattern in the programming behaviour of a specific group of programmers (e,g. novice, intermediate or expert programmers), one of the critical issues is the selection of similar participants who can be placed in one group. In this study we were interested in finding a method that could shorten the path to finding participants. Therefore, the use of clustering algorithms to group similar participants was put to test in order to investigate the effectiveness and feasibility of this approach. The clustering algorithms that were used for this study were K-means and DBSCAN. The results showed that the use of these algorithms, for the mentioned purpose, is feasible and that both algorithms can identify similar participants and place them in the same group while participants who are not similar to others, and therefore are not the correct subject of the study, are recognised.
机译：设计用于程序设计教育研究的实验时，需要仔细考虑以验证实验，在该实验中，需要收集和解释有关程序员的大量定性数据。当要在一组特定的程序员（例如，新手，中级或专业程序员）的编程行为中找到一种模式时，关键的问题之一就是选择可以放在一组中的相似参与者。在这项研究中，我们对寻找一种可以缩短寻找参与者的方法的方法感兴趣。因此，为了研究这种方法的有效性和可行性，对使用聚类算法对相似参与者进行分组进行了测试。用于本研究的聚类算法是K-means和DBSCAN。结果表明，出于上述目的，使用这些算法是可行的，并且两种算法都可以识别相似的参与者并将其放置在同一组中，而与其他人不相似的参与者则不是该参与者的正确对象。学习，得到认可。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号