巨型数据库中的数据采掘

数据采掘,也称数据库中的知识发现。传统进行数据分析的算法假设数据库中相关的记录比较少,然而,现在的许多数据库大到内存无法装下整个数据库,为了保证高效率,运用到大型数据库中的数据采掘技术必须是高度可缩放的。文章讨论了当今若干种先进的算法,它们能处理三类数据采掘:市场篮子分析、分类和聚类,并提出了今后的若干研究热点。

Data Mining in Very Large Databases

Data mining,also known as knowledge discovery in databases.Traditionally,algorithms for data analysis assume that the input data contains relatively few records.However,current databases are much too large to be held in main memory.To be efficient ,the data miming techniques applied to very large databases must be highly scalable.This paper introduces the characteristic of date-mining in very large database.The authors describe advanced algorithms that ad-dress three classical data-mining problems :market basket analysis,classification and clustering,and predict the popular problems in future.