一种非结构化数据的多牵度分配存取实现方法  

Method of Unstructured Data Access Based on Multi-level Assignment

作　　者：黄远鸣[1] 

机构地区：[1]广西经济管理干部学院计算机系,南宁530007

出　　处：《科技通报》2014年第8期161-163,共3页Bulletin of Science and Technology

基　　金：广西自然科学基金项目(2011GXNSFA018170)

摘　　要：非结构化数据随着信息时代的发展呈现爆炸式增长,非结构化数据存取的难点在于数据没有非常统一的规律便于系统查询和访问。传统的非结构化数据存取方法沿袭了结构化数据存取的方法,通过建立相应的特征关键字进行存取,由于非结构化数据的模糊特征性,对大量数据的兼容性很差。提出了一种非结构化数据的多牵度分配存取实现方法,将数据特征提取,根据特征将数据存储在不同的牵度区域,在数据合成时,通过牵度索引,建立不同的数据拼接顺序,从而实现很好的数据恢复。通过实际的海量数据进行测试实验,结果显示,采用多牵度分配存取的方法,非结构化数据的存取速度大大提高,具有很好的应用价值。With the development of the information age, the unstructured data presented explosive growth, and the difficult of unstructured data access was no uniform laws to facilitate the system queries and access. In traditional method, the un-structured data access method of a structure of data access was used, the access was achieved with the key by the establish-ment of appropriate characteristics, the blur characteristic of unstructured data, and when the data was large, the perfor-mance became poor. A kind of unstructured data access method was proposed based on multi-level assignment, through da-ta partitioning and re-engineering ideas. Tested experiment was carried out by actual experiment with massive data, the re-sults show that with the method of multi-access pull-degree distribution, the unstructured data access efficiency greatly im-proved, it has good value for application.

关 键 词：非结构化数据 多牵度分配 数据存取 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]