Skip to content

Latest commit

 

History

History
28 lines (22 loc) · 1.4 KB

README.md

File metadata and controls

28 lines (22 loc) · 1.4 KB

Resume-Matching

The follwing repository matches resume from various categories to given job descriptions.

Datasets:

Resume Dataset: Kaggle Resume Dataset

Job Descriptions: Job Description Dataset

Task Action

  • PDF Extraction: A PDF extraction tool is used to parse the PDF storing the metadata associated with it.
  • Job Description: Job Descriptions are chosen from the above mentioned dataset.
  • Candidate Job Matching: Given n job descriptions find the most relevant resume among available candidates.

Running the program

Download the resume datasets from the mentioned link.

Install the requirements.txt file. Python version used: 3.9.17

pip install requirements.txt

Run the following command

python ./src/resume_match.py -n 2 -w bottom

Flag -n (integer) specify the number of job descriptions to test against and -w (str) denote the location in dataset to choose job descriptions from, choices are [top, bottom, shuffle].

Note:

Parsing all the resumes (2484) may take from 10-20 minutes in an Intel-i5 CPU. 'test_resumes' folder contains one randomly sampled resume from each job category and can be used for quicker execution.