并行数据挖掘算法综述

Survey of Parallel Algorithms for Data Mining

数据挖掘在工业和商业领域中发挥着越来越重要的作用。随着数据量的增加,挖掘算法处理海量数据的能力问题日益突出。研究并行算法,是解决这一问题的有效途径。该文对常用的数据挖掘算法C4.5,SLIQ,SPRINT,关联规则,K-平均值,K-最近邻,贝叶斯网络,人工神经网络,遗传算法及并行性进行了研究探讨,为数据挖掘研究者提供借鉴。

Data Mining plays an important role in industry and business. The ability of data mining algorithms to deal with mass-data becomes more important with the increase in data. Parallel technology is an effective way to resolve this problem. This paper studies some common algorithms for data mining, such as C4.5, SLIQ, SPRINT, association rule, K-means, K-nearest neighbors, Bayesian network, artificial neural network and genetic algorithm, and their parallelism.