I am working on a resume parser project. Currently, I am using rule-based regex to extract features like University, Experience, Large Companies, etc.

So basically I have a set of universities' names in a CSV, and if the resume contains one of them then I am extracting that as University Name. In the same way I have a list of Large Companies in CSV and if the resume contains any of them then I flag it as Yes.

So these are rule-based logic and can never be fool-proof considering different countries have different resume formats. Is there any other way of doing it to improve the accuracy and make it a global solution?


1 Answer 1


Yes there is. Look at this paper of general information extraction

Where you can construct features and generalise around extracting, without any hard coded rules.

  • $\begingroup$ Thank you.. I will go through the document. Any sample code available for the same ? $\endgroup$
    – Akash
    Jan 17, 2020 at 9:52
  • $\begingroup$ No they did not, which I find bad but I guess its their product possibility $\endgroup$
    – Noah Weber
    Jan 17, 2020 at 9:57

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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