Leveraging GBM-XGBoost for Enhanced Skill-Based Job Role Recommendation Systems

Job role recommendation systems are essential for effectively connecting job searchers with possible employment prospects, particularly in aligning candidates' talents with job needs. Conventional methods, such Support Vector Machines (SVM) and Logistic Regression, demonstrate limitations in addressing non-linear connections and extensive, unbalanced datasets. This study presents a skill-oriented job position recommendation system employing the XGBoost model, a sophisticated gradient-boosted decision tree technique. The suggested system effectively manages extensive datasets, executes feature extraction from skill vectors, and enhances matching precision using a non-linear methodology. Experimental findings demonstrate that the suggested model attains an accuracy of 92.3%, markedly surpassing current methodologies, such as SVM (89.5%) and Logistic Regression (86.1%). The suggested model achieves an F1 score of 0.91 and a recall of 0.90, demonstrating its effectiveness in reducing false positives and false negatives. This study presents an innovative resolution to the issue of skill-based job matching, with applicable implications in recruiting, talent acquisition, and job recommendation systems. Future endeavours may concentrate on refining feature extraction techniques and integrating more layers in neural networks to further enhance scalability and accuracy.