• Version
• Download 39
• File Size 352.19 KB
• File Count 1
• Create Date 11 June 2024
• Last Updated 11 June 2024

A PROJECT ON RESUME DOMAIN CLASSIFICATION

MRIDUL DAYAMA, ADVIN MANHAR , PRAPHULLA MISHRA, MUKKU SUMANTH

^{AMITY  UNIVERSITY CHHATTISGARH, INDIA}

Abstract - In today's competitive job market, efficiently and accurately categorising resumes into appropriate job areas is critical for both recruiters and candidates. This study proposes a novel approach to resume domain categorization that makes use of advanced natural language processing (NLP) techniques and the K-Nearest Neighbours (KNN) algorithm. Our methodology makes use of numerous Python packages, including Sentence Transformer for embedding text into high-dimensional vectors, Docx2Txt and pyPDF for extracting textual data from various resume formats, and Pickle for model serialisation. Streamlit is utilised to create an interactive user interface, whereas Seaborn and NumPy are used for data visualisation and manipulation. The proposed system converts resumes to vector representations and uses the KNN algorithm to classify them into preset job areas. Extensive experiments show that our approach achieves great accuracy and robustness. This study not only advances the subject of automated resume processing, but it also offers a scalable solution for real-world applications in human resource management. The implementation details and performance evaluation demonstrate the potential for combining machine learning techniques with modern NLP technologies to improve resume analysis and classification.