对几种倒排文件压缩技术的研究与分析  

Research and Analysis on Five Inverted Files Compression Techiques

作　　者：王虎[1] 王潜平[1] 

机构地区：[1]中国矿业大学计算机科学与技术学院,江苏徐州221008

出　　处：《计算机工程与应用》2006年第7期169-173,共5页Computer Engineering and Applications

摘　　要：要提高文件检索系统的性能,需要对倒排文件压缩技术进行研究与对比,以使文件检索系统在最大压缩比和最快的解压速度间寻找均衡,以到达最大的吞吐量。对Golomb,Eliasgamma,Eliasdelta,VariableByteEncoding和BinaryInterpolativeCoding五种压缩技术通过在Windows操作系统下对theTRECWallStreetJournalcollection文件系统进行存取与压缩,从CPU时钟周期角度来对比各个算法的压缩比、压缩与解压缩的时间和对文件的读取和查询所花费的时间,并对它们进行了一个综合的评测。To improve the throughput of Information Retrieval System,we need to research into inverted file compression techniques,to get the statuesque point between best compression ratio and the fastest decompression time. In Windows operation system,from the idea of taking down the clock periods of CPU,through searching in the TREC Wall Street Journal collection file system,five compression techniques,Golomb,Elias gamma,Elias delta,Variable Byte Encoding and Binary Interpolative Coding are studied.The effect on file size,file seek time,and file read time are all measured as decompression time.In the end an integrative evaluation of the five techniques is drawn out.

关 键 词：倒排文件 索引 压缩 文本搜索 

分 类 号：TP311.52[自动化与计算机技术—计算机软件与理论]