fork download
copy 
  1. #-*- coding:utf-8 -*-
  2.  
  3. """ A basic lexicographic Similarity algorithm ...part1
  4.     @author: WhiZTiM (Timothy Onogu)
  5. 	@copyright: (c) January, 2013
  6. 	@NOTE: This algorithm is developed as it is, with the hope
  7. 		that it will be useful and I will not be responsible
  8. 		for any failures if any arises...
  9. 	@contact: whiztim@whiztim.com
  10. """
  11. from __future__ import division
  12.  
  13. def split_on_delimeters(string, delimeter=":;. \"!,$()-_+=~'`"):
  14.         """splits string upon the occurence of any character in delimeter"""    
  15.         rtn = []
  16.         placeHolder = ""
  17.         for x in string:
  18.                 if(not x in delimeter):                 #if current iter is not a delimeter
  19.                         placeHolder = placeHolder + x
  20.                         continue
  21.                 if(placeHolder != ""):                  #Not allowed to append empty string
  22.                         rtn.append(placeHolder)
  23.                         placeHolder = ""
  24.                         continue
  25.         if(placeHolder != ""):                          #Add the last item that is not catered for in the "for-loop"
  26.                 rtn.append(placeHolder)
  27.         return rtn
  28. def returnbigrams(string):
  29.         """This function returns bigrams"""
  30.         return [string[n:n+2] for n in range(len(string) - 1)]
  31.  
  32. def word_similarity(word1, word2, case_sensitive=False):
  33.         """This function, returns in percentage, how similar the string 'word1' is to 'word2'
  34.            >..>case_sensitive for considering cases"""
  35.         if(not case_sensitive):
  36.                 word1 = word1.lower()
  37.                 word2 = word2.lower()
  38.         pairs_word1 = returnbigrams(word1)
  39.         pairs_word2 = returnbigrams(word2)
  40.         t = len(pairs_word1) + len(pairs_word2)
  41.  
  42.         sb = 0
  43.         for x in pairs_word1:
  44.                 for y in pairs_word2:
  45.                         if(x == y):                     #bigrams match
  46.                                 sb += 2.0               #add (1+1=2).. since its found in both.
  47.                                 pairs_word2.remove(y)   #we do not need it again
  48.                                 break                   #break inner loop
  49.  
  50.         similarity = (sb / t * 100.0)
  51.         return similarity
  52.  
  53. def get_weight(word1, words1):
  54.     w = round((word1 / words1), 3)
  55.     return w
  56.  
  57.  
  58. sent1= "Getting higher on rank oh"
  59. sent2= "Getting on higher rank"
  60.  
  61. list1 =  split_on_delimeters(sent1)
  62. list2 =  split_on_delimeters(sent2)
  63. list1_ratios=[]
  64. weight_list=[] #new array
  65. weighted_similarity_list=[] #new array
  66. sum1 = 0 #new array - used to get sum of words for weight
  67.  
  68. if len(list1) > len(list2):
  69.     for x in list1:
  70.         list2_ratios=[]
  71.         sum1+=len(x) #sum of words is incremented
  72.         for y in list2:
  73.             list2_ratios.append(word_similarity(x,y))
  74.         list1_ratios.append(max(list2_ratios))
  75.  
  76.  
  77. else:
  78.     for x in list2:
  79.         list2_ratios=[]
  80.         sum1+=len(x) #sum of words is incremented
  81.         for y in list1:
  82.             list2_ratios.append(word_similarity(x,y))
  83.         list1_ratios.append(max(list2_ratios))
  84.  
  85. print str(list1_ratios) + " - Similarity list"
  86.  
  87. #Getting weight list
  88. for x in list1:
  89.         weight_list.append( get_weight(len(x), sum1))
  90. print str(weight_list) + " - Weight List of words in LiST 1"
  91.  
  92. #Getting weighted similarity
  93. for x in range(len(weight_list)):
  94.         weighted_similarity_list.append( round((weight_list[x] * list1_ratios[x]),3))
  95. print str(weighted_similarity_list) + " - Weighted similarity  List of words in LiST 1"
  96.  
  97. #Getting similarity
  98. similarity = sum(weighted_similarity_list)
  99. print str(similarity) + " - Similarity"
  100.  
  101.  
  102.  
  103.  
  104.  
  105.  
Success #stdin #stdout 0.08s 10840KB
comments (?)
stdin
copy 
Standard input is empty
stdout
copy 
[100.0, 100.0, 100.0, 100.0, 0.0] - Similarity list
[0.333, 0.286, 0.095, 0.19, 0.095] - Weight List of words in LiST 1
[33.3, 28.6, 9.5, 19.0, 0.0] - Weighted similarity  List of words in LiST 1
90.4 - Similarity
https://ideone.com/qZyAvx
language:
Python (cpython 2.7.16)
created:
11 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!