APPROXIMATE-TEXT-MATCH

KO | EN

gitlite — search

#javascript #python #hacktoberfest #react #ai #typescript #llm #go #golang #android #machine-learning #rust #deep-learning #linux

APPROXIMATE-TEXT-MATCH

★ 13 Open GitHub ↗

APPROX EQUALS AND APPROX CONTAINS

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# FUZZY-STRING_EQUALS AND FUZZY STRING_CONTAINS

DEMO:

Instructions to run JAR file:
Downlaod the jar file SAMPLERUN.jar
Open the command line and enter the commmand
java -jar [path to the downloaded jar in your local system ].
INPUTS:
parameter1 - String1
parameter2 - String2
parameter3 - method name - co for approximately contains eq for approximately equals
parameter4 - threshold value - (0,1]. 0 is trivial case which returns match for every pair of strings

FUZZY EQUALS AND FUZZY CONTAINS
 
Implementation of a new algorithm to find if a string is exactly/approximately contained as well as exactly/approximately equals another string exploiting the levenshtein distance algorithm. Given a minimum similarity threshold by the user, the algorithm returns a boolean if the calculated similarity ratio between the two strings is greater than the threshold provided by the user.

CODE STRUCTURE :
* The class consists of 3 functions.

FUNCTION NAME - MAIN
*       param1 - String needle
*       param2- String haystack
*       param3- The function name "eq" for approximately equals and "co" for approximately contains
*       Param4- The threshold value for similarity (0,1] 1 - for exact match 
FUNCTION NAME - APPROXIMATE_EQUALS

*       param1 - String needle
     *       param2- String haystack
     *       Param3- The threshold value for similarity (0,1] 1 - for exact match
EXPLANATION :
     *       If needle length is greater than haystack length we swap the two values and continue.Checks if the string and needle are exactly equal. If yes , no matter any threshold, it returns true.If they are not equal but the threshold given is 1 which measns the user is expecting an exact match, then it checks if there is an exact match and prints the result.If the user enters any value between 0 and 1 for the threshold, then the function returns true if the similarity ratio > threshold and returns false if the similarity ratio is <threshold.
     *       Similarity ratio= 1-levenshtein edit distance/max(length(needle,haystack))
FUNCTION NAME - APPROXIMATE CONTAINS
     
     *       param1 - String needle
     *       param2- String haystack
     *       Param3- The threshold value for similarity (0,1] 1 - for exact match
EXPLANATION :
     *       If needle length is greater than haystack length we swap the two values and continue.Checks if needle is an exact substring of haystack. if yes it returns TRUE no matter any threshold value. If the user requires an exact match giving a threshold of 1 expecting an exact contains, then checks for exact substring condition and returns true or false. If the threshold is any value between 0 and 1 then the algorithm uses a sliding window of needle length across the length of the haystack and checks if the first character of the needle is present in the haystack.If the first character(prefix of length 1 character) is present in the haystack then it calculates the levenshtein distance and checks if the ratio of similarity is greater than the threshold to return true or false respectively.
     *       Similarity ratio= 1-levenshtein edit distance/max(length(needle,haystack))

APPROXIMATE-TEXT-MATCH

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README