引入宽度优先扩展概念的多样性数据分类过程

Introduction of Breadth-First Extension of the Concept of Diversity Data Classification Process

随着信息技术的迅猛发展,大型数据库中的数据正在呈几何级数增长,冗余分类技术已经成为信息组织的重要手段。在总结归纳已有的冗余分类算法弊端的基础上,参考决策树算法提出了一种新的基于数据库抽样的海量数据冗余分类算法。算法引入宽度优先扩展思路,并给出了算法的基本思想和具体实现过程,探讨了并行环境下进行计算的优化步骤。在与SPRINT算法的比较实验中,新算法展现了良好的性能和分类效率。

With the rapid development of information technology, large-scale data in the database is growing exponentially. Redundancy classification technology has become an important means of information organization. In summarizing the four existing redundant classification algorithm based on the reference paper proposes a new decision tree algorithm based on sampling of massive database data redundancy classification algorithms, and describes the basic idea of the algorithm and implementation steps to explore parallel computing environment optimized. SPRINT algorithm with the comparative experiments, the new algorithm is to show a good performance and classification efficiency.