String Matching Algorithm based on Letters' Frequencies of Occurrence
Majed AbuSafiya · 2018
In this paper, we utilize the varying frequencies of occurrence of letters in the words in natural languages to propose a new string matching algorithm. Unlike standard string matching algorithms that compares letters in the pattern against the text in fixed order, the proposed algorithm ranks the letters of the pattern according to their frequency of occurrence. The letters of patterns are then matched against text according to that ranking, starting with the letter with the least frequency of occurrence. The proposed algorithm significantly outperformed KMP algorithm in terms number of comparisons.