Compare texts and find grade of match

[AndreasN]

Hi there,

I have an excel file with two different columns filled with text and would like to measure by rows (i.e. A2 to B2; A3 to B3...) how much (in %) these texts match with each other based on common words (punctuation is not a matching criteria; when the same word begins with a capital letter or is fully written in capital letters or only in small letters this should result in a match).

Many thanks

[pedrodrfaria · Accepted solution]

Hi @AndreasN 

Using the Fuzzy Matching would be the closest thing regarding giving you a score between a text comparison.

However, you can build out a logic to compare with word by word and see if the match, sum up and compare to the total and get a percentage.

Besides this, I dont know if there is an easy way to calculate this.

Pedro.

[pedrodrfaria · Accepted solution]

@AndreasN 

I attached one example of creating a manual calculation and scoring both string values :

In the example below I did only one sentence.

[echuong1 · Accepted solution]

You can try using fuzzy match.

I started by pre-processing the data with a data cleansing to remove duplicate whitespace, and change everything to uppercase to make it case insensitive. I compared all values from the text field to itself, but you can do this with two separate lists as well.

I added a make group and find and replace to append the "normalized" value to the end, just in case you needed it.

[AndreasN]

Hi Pedro,

Many thanks for the tips, helped me a lot to understand how to tackle my work.

Best,

Andreas

[AndreasN]

@echoung1

Many thanks for the tips, helped me a lot to understand how to tackle my work.

Best,

Andreas

[echuong1 · Accepted solution]

If your question is answered, please mark this thread as solved so others can find answers more easily. Thanks!