Abstract:
With the accomplishment of digital recruitment, now it is easier to find a job and apply
for it. At the same time organizations tends to receive a high volume of resumes for
each job posting. Managing and processing these resumes to filter a suitable candidate
is a time-consuming and costly task for Human Resource department. On average single
hiring takes 30 to 40 days with an above 4000$ budget. Resumes contain information
in many formats. CVLyzer helps the recruiters by providing automation of ranking
resumes according to job descriptions in a flash of time without any biasness. The
system also generates a report of ranking explanations for every individual. Besides
that, CVLyzer also assist the job seeker by providing a tailored review of the resume
with suggestions for improvements. The project is mainly based on Natural Language
Processing a subfield of Artificial Intelligence. Information retrieval is done using
named entity recognition. The model is trained using Spacy which uses word
embeddings for its NER model which is multilayer CNN. The system also contains
some rule-based models and text processors to extract data.
