I've got a large dataset of > 10,000 Glassdoor plain text job descriptions, and I would like to get a machine learning model that is trained to identify a specific section in each job description. That is the Job Responsibilities (JR) part.
Have a look at this Google Document [login to view URL], and check the lines between 16 and 30.
I already have used regular expressions to catch the lines where JR starts and ends (typically, just after short-length lines that include words like Responsibilities, Duties, etc and immediately before the Qualifications / Requirements part). But people are people, and job descriptions are still written by them, hence there is a variability in the language used and I can't automatically identify all the JR sections in the full dataset.
So, if you think you can solve this challenge by developing some sort of an NLP-powered machine learning model, I'd love to hire you.
If hired, you will be provided with the dataset subset that I have managed to positively identify the JR part (it'll be sth like is: text, job_id, jr_1_line_number, jr_2_line_number), and the full dataset of course.
I guess you'll be happy to hear that the training subset contains > 4,000 job description text files.
Please, bid only if you've successfully done sth very similar in the recent past!
The deliverable will be the trained model (with details on accuracy, recision, recall), the results on the full dataset, and all the scripts that you used to make it happen.
I will award the most awesome bidder that can do the full job in Python 3!
Cheers and happy bidding!
36 фрилансеров(-а) в среднем готовы выполнить эту работу за £207
Hello, I have read the details provided and i am positive i can provide quality work,please contact me to discuss more on the project deadline and some other few things