A new split based searching for exact pattern matching for natural texts
Document Type
Article
Publication Date
1-1-2018
Abstract
Exact pattern matching algorithms are popular and used widely in several applications, such as molecular biology, text processing, image processing, web search engines, network intrusion detection systems and operating systems. The focus of these algorithms is to achieve time efficiency according to applications but not memory consumption. In this work, we propose a novel idea to achieve both time efficiency and memory consumption by splitting query string for searching in Corpus. For a given text, the proposed algorithm split the query pattern into two equal halves and considers the second (right) half as a query string for searching in Corpus. Once the match is found with second halves, the proposed algorithm applies brute force procedure to find remaining match by referring the location of right half. Experimental results on different S1 Dataset, namely Arabic, English, Chinese, Italian and French text databases show that the proposed algorithm outperforms the existing S1 Algorithm in terms of time efficiency and memory consumption as the length of the query pattern increases.
Keywords
Algorithms, Data Mining, Software
Divisions
fsktm
Funders
Fundamental Research Grant (FRGS) through the University Malaya under Project No. FP003-2016 and IPPP research fund (PG017-2015B)
Publication Title
PLoS ONE
Volume
13
Issue
7
Publisher
Public Library of Science