可扩展双向聚类算法在煤炭领域中的研究与实现

随着现代信息技术的发展,各行各业产生了大量的高维数据,用不同的属性描述数据。煤炭产业也产生了大量的高维数据。煤炭企业的管理者往往希望从这些海量高维数据中得到更多的隐藏价值的知识。双向聚类算法被广泛地应用在各个领域中,该算法能够准确地完成聚类。随着数据规模呈现指数级地增长以及数据维度的增加,传统双向聚类算法不仅不能快速完成数据聚类,而且不能有效地处理高维数据。文章针对海量高维数据,提出了可扩展的基于高维数据的分布式双向聚类算法。通过实验结果展示了文中提出的算法具有很好的聚类结果以及很高的加速比及可扩展性。

Research and Implementation of Scalable Bi-clustering Algorithm in Coal Area

With the development of modern information technology,it produces huge scale high dimensional data in different areas,and they describe the data with different attributes.In coal area,it also produces huge scale coal data.The manager of coal industry usually hopesto get more hidden valuable knowledge from this huge scale high dimensional data.Bi-clustering algorithm is widely applied into different areas,and it could complete clustering accurately.With the exponentially increment of data size and data dimension,traditional bi-clustering algorithm not only could not complete data clustering fast,but also could not deal with high dimensional data effectively.In this paper,focusing on huge scale high dimensional data,we propose scalable distributed co-clustering algorithm based on high dimensional data.Through the experimental results,it shows that the algorithm in this paper has good clustering result,high speed-up and good scalability.