vishnunelapati / dl-nlp-resume-evaluation

Deep Learning - Curriculum Vitae matching job description using Natural Language Processing & Computer Vision

Deep Learning - Applied Natural Language Processing & Computer Vision

Curriculum Vitae Matching Job Description

Project done by:

• José Cabral
• José Jesus
• Manuel Sousa

Description

This project was executed in order to apply deep learning techniques learned in classes and in a way mix them in a project format whose concept would enable to explore both areas of natural language processing and computer vision. The work developed stands out as a program that selects the curriculum vitaes with the best match classification according to a job announcement description. Furthermore, the keywords present in the best resumes are analyzed and the faces of each candidate are detected in their own curriculum vitae.

Before running, please install all required dependencies using the command on CMD console:

pip install -r requirements.txt

To run the program use the command:

python run.py

...and is running on your CMD console!

How to use the program:

1. Create necessary txt file named "job_description" and folder named "resumes". If they are not created before running the program, the program itself will create them for you;
2. Write a job description into job_description.txt file;
3. Import resumes(CVs) pdf files into /resumes directory;
4. Press Enter key to run the program logic or [q] key to exit;
5. The best scored resumes will be available in /results folder following the images of each candidates face detected;
6. In the CMD console the results summary will also be printed: resume name, algorithm matching score and relevant keywords found in resume.