Resume Parsing Techniques

  • Thread starter Thread starter Juan Romero
  • Start date Start date
J

Juan Romero

Hey guys,

Has any of you played with Resume Parsing?

I have a project that involves automatically parsing resumes into a
database.

If you have any ideas, suggestions, resources, links, etc, please let me
know.

Thanks!
 
Juan Romero said:
Hey guys,

Has any of you played with Resume Parsing?
Click to expand...

Resume parsing?! Is this some sort of "black art"? :) I have never heard
of such things, but one way to make it easier would be to force your
applicants to submit their resume via a web form, then just pull the info
in.
 
Yes my dear Jeremy, unfortunately, it is real. I have seen it once or twice
before, but didn't pay much attention to it. I have seen places where you
paste your resume as text, and they parse it automatically into the fields.
The problem is that resumes come in different flavors, so it becomes very
difficult to parse it.
As for having the users fill out the fields, the reason why I posted this
question here is that they DON'T want to make users fill it out, lol.

Thank you for replying, I appreciate your help.
 
That is exactly the problem. Although there are many "standard" formats for
resumes (like the templates you get in MS Word), there is no garantee that
users will create resumes following this format. Resumes are free-style, and
so you cannot simply rely on standard formats.
I have been playing with Regular expressions, trying to look for common
clues in a document that allow me to parse the information. I was hoping
someone else out there had worked on this before. I am interested in hearing
different approaches, with some samples and stuff.

Thanks for answering!
 
* "Juan Romero said:
That is exactly the problem. Although there are many "standard" formats for
resumes (like the templates you get in MS Word), there is no garantee that
users will create resumes following this format. Resumes are free-style, and
so you cannot simply rely on standard formats.
I have been playing with Regular expressions, trying to look for common
clues in a document that allow me to parse the information. I was hoping
someone else out there had worked on this before. I am interested in hearing
different approaches, with some samples and stuff.
Click to expand...

Why not provide a preformatted Word template which everybody must use
for a digital resume? I don't see any other solution which makes
sense...
 
Back
Top