Optimizing Job Recommendation Systems with AI: A Deep Dive into BERT and GPT Models

Ari, Kiymet Elif (2024) Optimizing Job Recommendation Systems with AI: A Deep Dive into BERT and GPT Models. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

The increasing volume of job postings on online platforms has created a need for efficient job recommendation systems that enhance user experience and streamline job searches. This research investigates the optimization of job recommendation systems using advanced natural language processing models, specifically BERT (Bidirectional Encoder Representations from Transformers) and GPT (Generative Pre-trained Transformer). I utilized a live dataset, which includes job descriptions, company details, and user interactions. I performed extensive data cleaning and preprocessing to normalize text data, encode categorical features, and engineer a Job Desirability Score. This score integrates normalized company ratings, review counts, and job type weights to quantify job attractiveness. BERT was utilized to generate semantic embeddings for job descriptions and capturing contextual nuances. GPT was utilized to model user preferences for personalized recommendations. The performance of both models was evaluated. BERT achieved a superior validation accuracy of 76.54% and an F1 score of 0.7532, and a balanced precision-recall ratio. In comparison, GPT achieved a validation accuracy of 70.37% and an F1 score of 0.6364, with greater fluctuations in validation loss. These results underscore the potential of BERT in providing more accurate and contextually relevant job recommendations. The results suggest that incorporating advanced NLP models like BERT significantly enhances job recommendation systems. Further improvements can be achieved through hyper-parameter tuning, larger datasets, and exploring other model architectures. This research provides a foundation for future developments in artificial intelligence driven job recommendation systems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Zahoor, Sheresh UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence H Social Sciences > HD Industries. Land use. Labor > Issues of Labour and Work > Labour Market Z Bibliography. Library Science. Information Resources > ZA Information resources > ZA4150 Computer Network Resources > The Internet T Technology > TK Electrical engineering. Electronics. Nuclear engineering > Telecommunications > The Internet
Divisions:	School of Computing > Master of Science in Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	17 Jun 2025 18:23
Last Modified:	17 Jun 2025 18:23
URI:	https://norma.ncirl.ie/id/eprint/7896

Actions (login required)

View Item