Matching Algorithm

The WizSame matching algorithm runs as follows:

Phase 1:

If there are N words in fieldA of record1, and M words in fieldA of record2, and N > M, then the two records are similar if –

(1) More than 50% of the words in fieldA of record1 (the record having more words) are “similar word” to words in fieldA of record2; or

(2) All the words in fieldA of record2 (the record having less words) are “identical” to the words in fieldA of record1

Where –

“identical” is NOT case sensitive, and the order of words is not taken into consideration

“similar word” is NOT case sensitive, and

if there are N1 characters in word1 and M1 characters in word2, and N1 > M1, then the two words are “similar”if word1 starts with word2

Phase 2:

In this phase WizSame applies the Similarity Algorithm on all the characters. The WizSame Similarity Algorithmis NOT case sensitive

The WizSame Similarity Algorithm declares similarity between string A and string B if –

(1) The two strings are identical, or

(2) The only difference between the strings is that –

(a) One character in string A is missing from string B

(b) One character in string B is missing from string A

(c) One character in string A is replaced by another character in string B

(d) the order of two adjacent characters (ab) in string A appears in a reverse order in string B (ba).