C# .NET - Resume Parser help needed - Asked By Harsha Potlapalli on 01-Feb-10 02:39 PM

Hi Experts,

I am building an application which parses the candidate's resumes/Cv (Like word docs, PDF, txt etc..) in my .net application. The goal here is to have as much accurate and meaningful information(for example : skills,candidate's address,experience,job history,etc) extracted by the application from the resume.

Please if anybody has worked on a similar application, Can you please provide me the guidance on how to extract the information from resume/Cv?. If any samples codes wld be of great help to me.

Looking forward for the experts reply. Thanks in advance.


Harsha




Peter Bromberg replied to Harsha Potlapalli on 01-Feb-10 07:56 PM

realistically, there is no such thing as "resume parsing". Everybody's resume is different, has a different layout, different headings and sections.

Unless you use form fields with names such as you would find on major job posting sites (Monster.com, CareerBuilder, etc) this will likely prove to be an exercise in futility.

Sakshi a replied to Harsha Potlapalli on 01-Feb-10 09:04 PM

As peter said everyone's resume will have different headers.

So you should write your code in such a way that it understands the content even if the title differs.

write your code to decide the title by getting idea after parsing and analyzing the content of the heading.


Thanks and Regards,
http://www.CodeCollege.NET
http://www.InterviewsGuru.info

Harsha Potlapalli replied to Sakshi a on 02-Feb-10 11:00 AM

Hi All,

So you mean i have to create a dictionary of the keywords and then use those keywords for parsing the whole resume?? Can you provide me some sample code if you have any.

Thanks & Regards,

Harsha




Sakshi a replied to Harsha Potlapalli on 02-Feb-10 11:02 AM

just told you an idea. i dont have any code. but u can simply write it.

just u need to put 2 loops one inside one.


Thanks and Regards,
http://www.CodeCollege.NET
http://www.InterviewsGuru.info

Harsha Potlapalli replied to Sakshi a on 02-Feb-10 01:11 PM

Ok, anyways thanks for your help (Sakshi) for giving me some startup ideas. I will find out a way of doing it.

Thanks too you to Peter for sharing your views.


Thanks & Regards,

Harsha

Vinay Johar replied to Peter Bromberg on 01-Apr-10 12:22 AM
Hello Peter

I am sorry to say -- Resume Parsing is a realistic picture today. Yes I agree every resume is a different but Fuzzy and Artificial Intelligence has really picked up good and system is there to identify the format and extract data with more than 95% accuracy.
you can see demo at http://resumeparser.rchilli.com   

Regarding original question how to write parser -- please contact us directly we will be glad to help you out in this.
thanks
Warm Regards
Vinay
CEO, RChilli
vinay@rchilli.com
Ankita replied to Vinay Johar on 17-Apr-10 01:18 AM
Extracting information from an unstructured document is not a rocket science but accuracy level depends on many a things. Parsing works on probability. To extract a particular information from any unstructured document be it a resume or any other document, we need to write regular expression; internet has a good collection of standard regular expressions to extract various fields like name, email, phone, address, zip code etc.. we then need to have a dictionary in which we will fill the probability criterion and validation dictionary which will be required to validate the parsing results. To increase the accuracy level of the parsed contents , one needs to continuously enhance the dictionary and fill it with words which are used most commonly by mass. Multiple loops are required to reduce the possibility of errors.
Kanchana replied to Harsha Potlapalli on 29-Jan-11 01:54 AM
We has to develop a resume parser as a project.. which has to parse the resume/CV that is in doc,txt,pdf etc.. and should extract useful information from tat..pls help us regarding how to write  a resume parser..

                                                Thanks in Advance.
Aravind replied to Harsha Potlapalli on 31-Mar-11 03:01 PM
I am also building an application which parses the candidate's resumes/Cv (Like word docs, PDF, txt etc..) in my .net application. The goal here is to have as much accurate and meaningful information(for example : skills,candidate's address,experience,job history,etc) extracted by the application from the resume.


if hav any sample code jst post me......
ur junior..........
Raghuveer replied to Harsha Potlapalli on 29-Apr-11 03:19 AM
Hi... Harsha. I m also working on same project which u have done... plz advice me how to create a resume Parser...
i u have a code then plz send me...


thank in advance.