基于外存的概念格维护算法  

Out-of-core algorithm of maintaining concept lattice

作　　者：王春月[1] 王黎明[1] 张卓[1] 

机构地区：[1]郑州大学信息工程学院,河南郑州450001

出　　处：《计算机工程与设计》2018年第3期701-709,共9页Computer Engineering and Design

基　　金：国家青年科学基金项目(61303044)

摘　　要：为有效解决超内存概念格的维护问题,提出一种有限内存空间下基于外存的概念格维护算法。以形式背景中二元关系消减情况为应用背景,利用概念簇为基本单位划分原始概念格,采用最近最少使用调度策略释放内存中无用的概念节点,从外存中调入所需的概念块,降低内存消耗,解决内存不足的现象,为概念格的具体应用提供保障。理论与实验结果表明,在限定的内存资源条件下,该方法以较小的内存消耗和时间代价,实现了超内存概念格的有效维护。To solve the maintenance of hyper-memory concept lattice problem effectively,a maintenance algorithm of concept lattice based on the external memory under a limited memory space was proposed.Based on the binary relation reduction of formal context,the original concept lattice was divided with the concept cluster as the basic unit.The useless formal concept was released from the memory using the LRU scheduling strategy,and the required conceptual blocks were transferred from the external memory,which effectively reduced the memory consumption and solved the phenomenon of insufficient memory and provided a specific application of the concept lattice.Both theoretical and experimental results show that under the condition of limited memory space,the proposed method achieves the effective maintenance of concept lattice with small memory consumption and time cost.

关 键 词：有限内存空间 二元关系消减 概念格 概念簇 外存算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]