The Resume parser was created in order to aid recruiters parse through large number of resumes in order to find those which have the required skills. This is not a complete tool but an aid to help shortlist resumes.
The single notebook allows the data to focus in one resume while the multiple resume parser extracts from a folder, which has to be provided.
pip3 install pdfminer
pip3 install spacy
pip3 install nltk
pip3 install pandas
I was largely inspired by the methods and techniques used by Omkar Pathak for more details - https://omkarpathak.in/2018/12/18/writing-your-own-resume-parser/
From the notebook, you can see how much harder it is to extract data from PDF rather than words and the accuracy of extraction is also much lower. I think this is a possible place of improvement
For better classification increase the number of skills in the skills.csv file under the appropriate headers