Suppose I want to check if a string has 'pizza' in it. But my user accidentally typed 'I want piztza' by accident (a simple typo). Is there a way to check if the user input approximately contains 'pizza'? Perhaps if only off by one letter.
4 Answers
Try using fuzzywuzzy from PyPI: https://pypi.org/project/fuzzywuzzy/. I am not very experienced with it, but the website should help.
3 Comments
alani
You would probably use
fuzz.partial_ratio. See my comments under the question for an example.
Krishnan Shankar
You should just post your answer as an answer to the question instead of a comment.
alani
The culture here is definitely that answers need to be somewhat more developed than what gets posted as comments. Mine was somewhat borderline, but I am happy to just leave it as a comment, especially in view of other answers that have since appeared. Yours, if you don't mind me saying, fits firmly into the category of what most people would post as a comment rather than an answer.
fuzzywuzzy would be one way to do this. fuzzywuzzy library compares the similarity between two strings. The Jaccard distance is one method that fuzzywuzzy uses. The Jaccard distance - measure any two sets.
This is what the formula looks like :
Here is a visual representation of the formula:
Here is how you could implement it:
from fuzzywuzzy import fuzz
similarity = fuzz.ratio("hello","world") # 0%
This will give you a ratio of how similar the words are. Here is another example:
from fuzzywuzzy import fuzz
similarity = fuzz.ratio("Austria","Australia") # Will give you 88%
Comments
you can use fuzzywuzzy package
from fuzzywuzzy import fuzz
q1 = input("type 'pizza': ")
compare = fuzz.ratio(q1, 'pizza')
if compare > 90:
print("you type pizza")
else:
print("you didn't type 'pizza'")
check his documentation for more uses, this is only a simple code.
1 Comment
ktyd
Wow thanks so much everyone! I checked out fuzzywuzzy but it seemed like it was more useful for directly comparing two similar strings and not for finding strings in strings. I think I will use something similar to what user13914826 posted. Thanks again!
Try this:
Sentence = "I want piztza"
Word = "pizza"
Same = False
for Sen_Word in Sentence.lower().split(" "):
if Sen_Word == Word:
Same = True
elif len(Sen_Word) == len(Word)+1:
Left_Count=0
Right_Count=0
for i, x in zip(Sen_Word, Word):
if i == x:
Left_Count += 1
else:
break
Letters_Left = min(len(Sen_Word), len(Word))-Left_Count
for i, x in zip(Sen_Word[::-1], Word[::-1]):
if Letters_Left - Right_Count > 0 and i == x:
Right_Count += 1
else:
break
if Left_Count + Right_Count == len(Word):
Same = True
print(Same)
or a shorter code:
Sentence = "I like piztza"
Word = "pizza"
Same = False
for Sen_Word in Sentence.split(" "):
if Sen_Word == Word: Same = True
elif len(Sen_Word) == len(Word)+1:
Left_Count=Right_Count=0
for i, x in zip(Sen_Word, Word):
if i == x: Left_Count += 1
else: break
Letters_Left = min(len(Sen_Word), len(Word))-Left_Count
for i, x in zip(Sen_Word[::-1], Word[::-1]):
if Letters_Left - Right_Count > 0 and i == x: Right_Count += 1
else: break
if Left_Count + Right_Count == len(Word): Same = True
print(Same)
Comments
fuzzywuzzypackage.from fuzzywuzzy import fuzzthenfuzz.partial_ratio("I want piztza", "pizza")gives 80, whereasfuzz.partial_ratio("I want fish", "pizza")gives 22. You would need to decide on an appropriate cutoff value.