![](https://github.com/karthagokul/pyresumize/blob/main/logo.png?raw=true)
Introduction
pyresumize is a python module to extract useful information from resume and generate a json string out of it. Currently it supports only pdf,docx files as input .
Todo
- Proper Logging to be added
- Support for other formats
- Performance Improvements
- Bug Fixes
- Custom configuration of input data
Notes
The Skills , Employers and Education is given as .csv inputs to the engine and you can see a reference implementation in the data folder.
Design
Class Diagram
![Class Diagram Alt text](https://github.com/karthagokul/pyresumize/blob/main/classes_pyresumize.png?raw=true)
Package Diagram
![Packages Alt text](https://github.com/karthagokul/pyresumize/blob/main/packages_pyresumize.png?raw=true)
I have changed the Design in such a way that the developers can create own parsing rules and set those to the Parser to bring in flexibility.
Currently we have the below interfaces exposed and the developers can override the process method to bring in custom processing rules.
Searches for company information in the resume and identifies the employment information from resume. The builtin module uses NER model from a set of 1500+ employers . I already have a plan to list all the employers based on countries (Future Releases)
Process Education Details together with universities .Similarly the data is very primitive , I need to update the input csv files based on countries(future releases)
Check for email addresses in the resume and returns if found one.
Process phone numbers in the resume text, if there are more than one phone number found , returns a concatenated string with commas.
Proces the Name of the candidate .
Process the skills section . returns a list of identified skills in a resume. I used a linked in scrapped input for this module.
Sometimes you will be in a need to extend there interfaces , Then create a class like below to have custom parsing.
class RemoteCompaniesChecker (EmployerBaseInterface):
def process(self,resumetext):
#Call a remote API and pass the text info
return list[]
The ResumeEngine class has below member functions and with one of these you can apply your custom engine
set_skills_engine(self, engine):
set_name_engine(self, engine):
set_name_engine(self, engine):
set_email_engine(self, engine):
set_education_engine(self, engine):
set_employer_engine(self, engine):
Currently the Engine supports PDF and DOCX File extensions . Incase if you want to use a different parsing mechanism for a particular type of file, you can extend AbstractFileProcessor and use add_custom_text_processor method to register your own text processing features.
Usage
https://pypi.org/project/pyresumize/ . The module can be installed using
pip install pyresumize
Then do the below
python -m spacy download en_core_web_lg
python -m nltk.downloader words
python -m nltk.downloader stopwords
from pyresumize import ResumeEngine
r_parser=ResumeEngine()
r_parser.set_custom_keywords_folder("data")
json=r_parser.process_resume(file)
print(json)