ijarce.org

Journal Archive

    1. Number 1
    1. Number 2
    2. Number 1

NEWS AND EVENTS

29 November 2016

Attention to Authors

The Volume 1, Number 2, November 2016 issue of IJARCE has been published and hardcopies have been sent to authors. Notice that, arrival process of the hardcopies, to your addresses, will takes between 12 to 30 days depending on your country of origin.

 

10 October 2016

Attention to Authors

The Volume 1, Number 1, September 2016 issue of IJARCE has been published and hardcopies have been sent to authors. Notice that, arrival process of the hardcopies, to your addresses, will takes between 5 to 30 days depending on your country of origin.

- -

Algorithms for Approximate String Matching

Artical 17 ; volume1 ; number2 ; Page: 73-77 ; PDF Download

Authors

Zafar Ali

Department of Computer Engineering, Federal Government University, Virtual University of Pakistan, Pakistan.

Abstract

String matching plays a major role in our day to day life be it in word processing, signal processing, data communication or bioinformatics. Approximate string matching is a variation of exact string matching that demands more complex algorithms. As the name suggests, in approximate matching, strings are matched on the basis of their non-exact similarities. Quantification of the parameter “similarity” and its efficient computation received a lot of attention in the recent past. Many definitions of similarity have been given depending on application needs. One of the standard ways of quantifying similarity is using the well-known edit distance. It is defined as the number of edits (insertions, deletions, and substitutions) needed to go from one string to another. If similarity or distance has a fuzzy co-efficient where “Hat” and “Fat” are considered the closest words due to the adjacency of the alphabet keys H and F to G. In a similar way, fuzzy concepts can be applied for a DNA sequence similarity where evolution is rule based. Many algorithms based on binary tree or improvement on it as suffix tree, dynamic programming, indexing of strings before searching are being used. Neural network gives a very good result as it is efficient to learn and eliminate case based errors. Further a fusion of fuzzy sets at the input, output and the neurons can greatly improve the matching process.

Statistics

Article View : 1450

PDF Download : 628

Keyword