The Resume Matcher is a Python-based project designed to match job descriptions with candidate resumes using natural language processing techniques. This tool aids in identifying the most suitable candidates for specific job roles by calculating the similarity between job descriptions and candidate resumes.
The project is divided into four main steps:
-Preprocess and tokenize both resumes and job descriptions. -Ensure consistent text formatting and language handling.
-Generate word embeddings for both resumes and job descriptions. -Utilize advanced models like DistilBERT for embeddings.
-Calculate cosine similarity between job descriptions and resumes. -Rank CVs based on similarity scores and list the top candidates.
The project directory structure is organized as follows:
The matching results are stored in the output/ directory in JSON format. You can explore the top candidates for each job description based on similarity scores.