A simple resume parser used for extracting information from resumes

Built with ❤︎ and ☕ by Omkar Pathak

• Extract name
• Extract email
• Extract mobile numbers
• Extract skills
• Extract total experience
• Extract college name
• Extract degree
• Extract designation
• Extract company names

• You can install this package using

pip install pyresparser

• For NLP operations we use spacy and nltk. Install them using below commands:

# spaCy
python -m spacy download en_core_web_sm

# nltk
python -m nltk.downloader words
python -m nltk.downloader stopwords

Official documentation is available at: https://www.omkarpathak.in/pyresparser/

• PDF and DOCx files are supported on all Operating Systems
• If you want to extract DOC files you can install textract for your OS (Linux, MacOS)
• Note: You just have to install textract (and nothing else) and doc files will get parsed easily

• Import it in your Python project

from pyresparser import ResumeParser
data = ResumeParser('/path/to/resume/file').get_extracted_data()

For running the resume extractor you can also use the cli provided

usage: pyresparser [-h] [-f FILE] [-d DIRECTORY] [-r REMOTEFILE]
                   [-re CUSTOM_REGEX] [-sf SKILLSFILE] [-e EXPORT_FORMAT]

optional arguments:
  -h, --help            show this help message and exit
  -f FILE, --file FILE  resume file to be extracted
  -d DIRECTORY, --directory DIRECTORY
                        directory containing all the resumes to be extracted
  -r REMOTEFILE, --remotefile REMOTEFILE
                        remote path for resume file to be extracted
  -re CUSTOM_REGEX, --custom-regex CUSTOM_REGEX
                        custom regex for parsing mobile numbers
  -sf SKILLSFILE, --skillsfile SKILLSFILE
                        custom skills CSV file against which skills are
                        searched for
  -e EXPORT_FORMAT, --export-format EXPORT_FORMAT
                        the information export format (json)

• If you are running the app on windows, then you can only extract .docs and .pdf files

The module would return a list of dictionary objects with result as follows:

[
  {
    'college_name': ['Marathwada Mitra Mandal’s College of Engineering'],
    'company_names': None,
    'degree': ['B.E. IN COMPUTER ENGINEERING'],
    'designation': ['Manager',
                    'TECHNICAL CONTENT WRITER',
                    'DATA ENGINEER'],
    'email': '[email protected]',
    'mobile_number': '8087996634',
    'name': 'Omkar Pathak',
    'no_of_pages': 3,
    'skills': ['Operating systems',
              'Linux',
              'Github',
              'Testing',
              'Content',
              'Automation',
              'Python',
              'Css',
              'Website',
              'Django',
              'Opencv',
              'Programming',
              'C',
              ...],
    'total_experience': 1.83
  }
]

• Some of the core concepts behind the algorithm have been taken from https://github.com/divapriya/Language_Processing which has been summed up in this blog https://medium.com/@divalicious.priya/information-extraction-from-cv-acec216c3f48. Thanks to Priya for sharing this concept

• https://www.kaggle.com/nirant/hitchhiker-s-guide-to-nlp-in-spacy

• https://www.analyticsvidhya.com/blog/2017/04/natural-language-processing-made-easy-using-spacy-%E2%80%8Bin-python/

• Special thanks to dataturks for their annotated dataset

If you have found my softwares to be of any use to you, do consider helping me pay my internet bills. This would encourage me to create many such softwares 😄

PayPal
₹ (INR)