Career Compass: Navigating to the Right Job with Machine Learning Precision
Chandana H M1, Raksha D2, Suhas H3, Sumanth B J4, Sunitha P5
1Assistant Professor of Computer Science and Engineering Department, Malnad College of Engineering, Hassan, Karnataka
2Student of Computer Science and Engineering Department, Malnad College of Engineering, Hassan, Karnataka
3Student of Computer Science and Engineering Department, Malnad College of Engineering, Hassan, Karnataka
4Student of Computer Science and Engineering Department, Malnad College of Engineering, Hassan, Karnataka
5Student of Computer Science and Engineering Department, Malnad College of Engineering, Hassan, Karnataka
1hmc@mcehassan.ac.in, 2rakshad2005@gmail.com, 3suhasharisha@gmail.com, 4sumanthdev9656@gmail.com, 5sunisuno2004@gmail.com
Abstract - Job recommendation systems are critical tools in bridging the gap between job seekers and employers by automatically suggesting suitable opportunities. This paper explores a resume- and skill-based job recommendation system using a hybrid approach combining machine learning, deep learning, and NLP techniques. The proposed system parses user resumes to extract structured skill sets and work experience using NLP. Then, it applies transformer-based models like BERT for contextual embedding of resume and job descriptions to calculate similarity scores.
Additionally, knowledge graphs are employed to improve the semantic understanding of skills and job roles. Through an extensive literature review and methodology analysis, this paper highlights state-of-the-art techniques, implementation frameworks, and evaluation metrics for developing a robust job recommender system. The results demonstrate that deep learning and contextual embeddings outperform traditional keyword-based methods in precision and personalization, offering a promising direction for intelligent recruitment platforms.
Key Words: Job Recommendation, Resume Parsing, BERT, Knowledge Graphs, Machine Learning, NLP, Transformer Models
