基于Hadoop的数据挖掘算法在葡萄酒信息数据分析系统中的应用

发布时间：2018-03-21 22:38

本文选题：Hadoop　切入点：数据挖据算法　出处：《计算机应用》2017年S1期 　论文类型：期刊论文

【摘要】：针对宁夏贺兰山东麓葡萄酒庄在销售自产葡萄酒的过程中存在葡萄酒信息数据分析不准确、销售渠道窄、销售信息更新慢、销量不高等问题,提出了一种基于Hadoop分布式框架的数据挖掘算法,对极大量的葡萄酒信息数据分析系统的数据进行采集、分析,并根据分析结果为用户推荐适合其口味的葡萄酒饮品。该系统以宁夏贺兰山东麓葡萄酒庄为实验基地,自主开发葡萄酒信息数据分析系统,并利用数据挖据算法中的聚类算法K-means算法和分类算法中的C4.5算法进行数据挖据;其中针对聚类算法中的K-means算法和分类算法中的C4.5算法实现分布式研究,实现了基于Map Reduce的分布式聚类和分类算法,并且在此基础上对其各自的缺点进行相应的改进。实验证明了基于Hadoop框架的数据挖掘算法在算法的稳定性和准确率上均有很好的提升,并且可以对葡萄酒信息数据分析系统的海量数据进行很好的处理。
[Abstract]:In view of the problems of inaccurate analysis of wine information data, narrow sales channels, slow updating of sales information and low sales volume in the process of selling wine from the eastern foothills of Helan Mountain in Ningxia, the wine farms have some problems, such as inaccurate analysis of wine information data, narrow sales channels, and so on. A data mining algorithm based on Hadoop distributed framework is proposed to collect and analyze the data of a large number of wine information data analysis system. Based on the results of the analysis, the authors recommend the wine drinks suitable for their taste. The system takes the eastern foothills of Helan Mountain in Ningxia as the experimental base and independently develops the wine information and data analysis system. The data mining algorithm is based on the clustering algorithm (K-means) and the classification algorithm (C4.5), in which the K-means algorithm and the C4.5 algorithm of the clustering algorithm are used to realize the distributed research. The distributed clustering and classification algorithm based on Map Reduce is implemented. On the basis of this, the corresponding improvement is made on their respective shortcomings. The experiment proves that the data mining algorithm based on Hadoop framework has a good improvement in the stability and accuracy of the algorithm. And the wine information data analysis system can do a good deal with the massive data.
【作者单位】：宁夏大学信息工程学院;
【基金】：宁夏大学研究生创新项目(GIP201625)
【分类号】：F426.82;TP311.13

【相似文献】