摘要
针对如何分析校园无线网络数据、挖掘数据中蕴藏的学生行为,更好地辅助教学管理,提出了在Hadoop平台构建基于自组织神经网络的模糊C-均值聚类算法。该算法采用自组织神经网络与模糊C-均值聚类算法相结合,避免了模糊C-均值聚类算法初始化不当带来的误差。考虑到无线用户网络行为数据规模庞大,采用了Hadoop平台并行运行聚类算法,有效地降低了分析时间。通过采集用户校园无线网络数据,利用聚类算法评估了学生群体的学习兴趣度。实验结果表明,提出的算法提高了聚类结果的准确性,分析平台为学校管理层有效地作出决策提供了依据,研究方法为其他高校解决类似问题提供了有益的参考思路。
Aiming at how to analyze the campus wireless network data,mine the student behavior behind the data,and better assist teaching management,this paper proposed a fuzzy C-means clustering algorithm on Hadoop platform,which was based on self-organizing neural network.The algorithm combined self-organizing neural network with fuzzy C-means clustering algorithm,which could avoid the error of improper initialization.Taking the huge scale of wireless user data into account,it adopted the Hadoop platform to execute the algorithm in parallel,which effectively reduced the analysis time.Through collecting user campus wireless network data,this paper used clustering algorithm to evaluate the degree of students learning interest.The experimental results show that the proposed algorithm can improve the accuracy of clustering results,and the analysis platform can provide basis for the school management to make decisions quickly and efficiently.The research method can provide a useful reference for similar questions for other colleges and universities.
作者
王法玉
姜妍
Wang Fayu;Jiang Yan(Tianjin Key Laboratory of Intelligence Computer&Novel Software Technology,Tianjin University of Technology,Tianjin 300384,China)
出处
《计算机应用研究》
CSCD
北大核心
2018年第1期186-189,共4页
Application Research of Computers
基金
天津市自然科学基金资助项目(15JCTPJC60100)
