A self-adaptive job recommendation system for candidates to find the best matching jobs.
Environment Requirement: Refer to the environment.txt file.
Cold start (without data):
- In the project root directory, run $python ./scripts/rating.py, you will be asked to input job rating to start, be patient and finish your data inputs (This is very important for the system to function properly in later steps.)
- After finishing initial data labeling, under the same project root directory, run $python ./scripts/resume_parser.py to parse the resumes into computer legible data formats.
- In the same directory, run $python ./scripts/pproducequery.py to generate query from parsed resume data.
- In the project root directory, run $python ./scripts/train_test_validation_generate.py to generate the datasets for further training and processing
- In the project root directory, run $python run.py Please be patient as the initialization time for a new user can take 2-3 mins
Normal start: Start with non-empty job_user_rating.csv: In project root directory, run $python run.py Please be patient as the initialization time for a new user can take 2-3 mins... Then follow the prompts on screen