Designing an Accurate and Efficient Algorithm for Matching Arabic Names

Soft Bigram distance for names matching

“…Due to the absence of standard datasets for name matching, different multilingual datasets (English, Arabic, Portuguese) is used in the experiments carried out in this research. These datasets are presented by (Al-Hagree et al, 2019a;Al-Hagree et al, 2019b;Ahmed & Nürnberger, 2009;Rees, 2014;Al-Sanabani & Al-Hagree, 2015). Different spelling errors and typographical are included in these datasets.…”

“…The same data preparation used in Al-Hagree et al (2019a); Al-Hagree et al (2019b)) is used in this research to be compared with the Soft-Bidist algorithm.…”

“…Generally, it is not easy to provide accurate weights between pair source and target. In order to gain proper values for weights wt1, wt2, wt3, wt4, wt5, wt6, wt7, wt8, and wt9, the 1 0,1,1,0,1,1,1,1 and 1 0.83 2 0,1,0,0,1,1,1,1 and 1 0.87 0,1,1,0.5,0.5,1,1,1 and 1 0.82 4 0,1,0,0.5,0.5,1,1,1 and 1 0.89 5 0,1,1,0,0.5,1,1,1 and 1 0.87 6 0,1,0,0,0.5,1,1,1 and 1 0.91 7 0,1,0,0.2,0.2,1,1,1 and 1 0.91 8 0,1,0,0.1,0.1,1,1,1 and 1 0.93 9 0,1,0,0,0.2,1,1,1 and 1 0.93 10 0,1,1,0,0.5,1,1,0.5 and 0.5 0.89 11 0,1,0.5,0,0.5,1,1,0.5 and 0.5 0.91 12 0,1,0,0,0.5,1,1,0.5 and 0.5 0.93 13 0,1,0.5,0,0.5,1,1,0.5 and 0.5 0.88 14 0,1,0,0.2,0.2,1,10.5 and 0.5 0.94 15 0,1,0,0.1,0.1,1,1,0.5 and 0.5 0.96 16 0,1,0.5,0,0.,1,1,0.5 and 0.5 0.95 17 0,1,0,0,0.2,1,1,0.5 and 0.5 0.96 18 0,1,0,0,0.1,1,1,0.5 and 0.5 0.97 experiments with different weights for Tables 1 and 2 of dataset 1 (Al-Hagree et al, 2019a;Al-Hagree et al, 2019b) should be repeated. The results are presented in Table 3.…”

“…This work concentrated on improving an algorithm to Identify Confusable Drug Names, based on Bi-gram algorithms and the blend of the longest common subsequences. Furthermore, the research work achieved (Al-Hagree et al, 2019a;Al-Hagree et al, 2019b) proposed an enhanced N-DIST method that concentrated on improving an algorithm for Name Matching. However, the previous studies differ from the contribution in this paper, because the proposed algorithm in this paper combines a Bi-gram technique with a distance technique (Al-Hagree et al, 2019a;Al-Hagree et al, 2019b).…”

“…To increase the accuracy of identifying the names, there is a need to find the set of weights WT = {wt1; wt2; ...; wt9} of the distance scale of Soft-Bidist. For this, a randomized value is used (Levenshtein, 1966;Al-Hagree et al, 2019a;Al-Hagree et al, 2019b;Kondrak, 2005;Millán-Hernandez et al , 2019;Rees, 2014).…”