5
$\begingroup$

This is my first time to use Data Analytics tool to figure out a solution to a problem. I have a table with following columns Person ID, Person Name, Note ID, Note (notes is a free form text where a call representative can enter their comments) Insert User. I have been given list of key phrases to be identified from the Note column. These key phrases are in a sentence format. Example:enter image description here

I have to find exact or a similar matching phrase from those notes. The format of my final report would look something like this enter image description here

I looked into couple of articles related to text matching which suggested options such as

fuzzywuzzy, Doc2vec, Difflib, python-levenshtein

It is all so confusing. Even if I get a starter to pick the best suitable option, I can may be take it from there. Any suggestions? Thank you so much!

$\endgroup$
1
  • $\begingroup$ can you upload the json text data than image $\endgroup$
    – Naga kiran
    Oct 1, 2018 at 10:52

1 Answer 1

1
$\begingroup$

The problem you are suggesting is a text summarization problem. It can be of two types abstractive - understanding the text distribution and then producing a summary, extractive - extracting specific words from the text to produce a summary. Clearly, your problem falls in the latter category. Also, since you have the labels to be generated it is a supervised text summarization problem. Take a look at this paper regarding algorithms for the same. It suggests two solutions for a similar problem, you'll of course have to modify them a bit and see which one works best for you. Good luck.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.