A recommender system for employee recruitment

• Main Author: Yeo, Boon Hao
• Other Authors: Josephine Chong Leng Leng
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Computer and Information Science
• Online Access: https://hdl.handle.net/10356/181186
• Institution: Nanyang Technological University

In today’s competitive job market, the increasing volume of resumes presents significant challenges for recruiters. These challenges include the time-consuming process of manual resume screening, difficulty in identifying qualified candidates due to lack of domain knowledge and risks of human error. This study investigates the use of machine learning, deep learning and natural language processing (NLP) techniques to automate resume screening and address these challenges. Models such as K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN) were evaluated, focusing on the impact of hyperparameter tuning, feature selection and dataset size on performance. Additionally, techniques like Named Entity Recognition (NER) for skill extraction and sentence transformers for semantic similarity comparison were explored. The use of sentence transformer models significantly improved the resume screening process by capturing contextual meaning and enhancing semantic comparisons between resumes and job descriptions. This resulted in more accurate and efficient identification of candidates whose qualifications closely align with job requirements. Feature selection, emphasizing attributes such as job titles, skills, and education, further increased classification accuracy. The study concludes that machine learning and deep learning models can effectively automate resume classification, improving recruitment efficiency and saving time.