We want to build a software application that understands resumes and produces results that is as accurate as a human result.
NLTK library should be used to retrieve words that are similar to the job description so that the program is able to process these keywords as part of something that the employers want.
Machine learning using SCIKIT-LEARN package should be used to classify these results and whether such keywords should be considered in future resumes.
The program should also be able to understand and store new keywords and phrases and subsequently store these phrases into a dictionary. Eg. if the employer is looking to hire an accountant, all words and phrases associated with programmer should be taken into account using the dictionary in NLP.
A phrase such as "I am able to speak 3 different languages, english, french, spanish" should also be taken into account as softskills of the candidate.
Sample resumes would be provided to feed the machine with new data.
The program should be coded in python and machine learning algorithms should be using scikit-learn.
Only a simple GUI is needed for employers to view different resumes and display the qualities of the recommended resume by the system. Eg. if the employer key in a job description such as "computer programmer" the system should display a rating of these resumes that it recommends.