A Gradient-Enhanced Decision Tree and XGBoost-Based Human-Job Matching Model

In recent years, there have been significant advancements in the field of natural language processing related to the extraction of features from resumes. Despite this progress, challenges such as job market indicators and information asymmetries persist. To address these challenges, this study introduces a person-post matching model that leverages a model fusion approach and the Multinomial Naive Bayes algorithm to convert text data into vector form, producing a sparse matrix. The resulting matrix is used for label classification by various classifiers, and the prediction outcomes of multiple base classifiers are combined using meta-classifiers. By taking advantage of the ensemble capabilities of the base classifiers and the meta-classifier, the model's performance is improved by 12%. Extensive experiments using real-world resume datasets confirm the effectiveness of our proposed person-job matching model.