期刊名称：International Journal of Emerging Technologies in Learning (iJET)
印刷版ISSN：1863-0383
出版年度：2015
卷号：10
期号：8
页码：55-58
语种：English
出版社：Kassel University Press
摘要：In the development of information technology the development of scientific theory has brought the progress of science and technology. The progress of science and technology has an impact on the educational field, which changes the way of education. The arrival of the era of big data for the promotion and dissemination of educational resources has played an important role, it makes more and more people benefit. Modern distance education relies on the background of big data and cloud computing, which is composed of a series of tools to support a variety of teaching mode. Clustering algorithm can provide an effective evaluation method for students' personality characteristics and learning status in distance education. However, the traditional K-means clustering algorithm has the characteristics of randomness, uncertainty, high time complexity, and it does not meet the requirements of large data processing. In this paper, we study the parallel K-means clustering algorithm based on cloud computing platform Hadoop, and give the design and strategy of the algorithm. Then, we carry out experiments on several different sizes of data sets, and compare the performance of the proposed method with the general clustering method. Experimental results show that the proposed algorithm which is accelerated has good speed up and low cost. It is suitable for the analysis and mining of large data in the distance higher education.
其他摘要：In the development of information technology the development of scientific theory has brought the progress of science and technology. The progress of science and technology has an impact on the educational field, which changes the way of education. The arrival of the era of big data for the promotion and dissemination of educational resources has played an important role, it makes more and more people benefit. Modern distance education relies on the background of big data and cloud computing, which is composed of a series of tools to support a variety of teaching mode. Clustering algorithm can provide an effective evaluation method for students' personality characteristics and learning status in distance education. However, the traditional K-means clustering algorithm has the characteristics of randomness, uncertainty, high time complexity, and it does not meet the requirements of large data processing. In this paper, we study the parallel K-means clustering algorithm based on cloud computing platform Hadoop, and give the design and strategy of the algorithm. Then, we carry out experiments on several different sizes of data sets, and compare the performance of the proposed method with the general clustering method. Experimental results show that the proposed algorithm which is accelerated has good speed up and low cost. It is suitable for the analysis and mining of large data in the distance higher education.
