Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 2 min ago
Peer-reviewedCC BY 4.0PLOS One

TPE-optimized XGBoost model for fine–grained prediction of tertiary vocational college students’ employment destinations

  • Juntao Chen1
  • Jinmei Zhan2
  • Xiaodeng Zhou3
  • Jiahua Yao4
  • Su-Kit Tang5
  • Xia Liu
5 affiliations
  1. 1School of Financial Management, Hainan College of Economics and Business, Haikou, China
  2. 2Faculty of Applied Sciences, Macao Polytechnic University, Macao, China
  3. 3School of Information Science and Technology, Qiongtai Normal University, Haikou, China
  4. 4Faculty of Humanities and Social Sciences, Macao Polytechnic University, Macao, China
  5. 5Sanya Aviation and Tourism College, Sanya, China
Open at sourcePDFDOICite

Abstract

Predicting tertiary vocational college students’ employment destinations has become an urgent need for most tertiary vocational colleges. This study proposes a machine learning framework based on Tree-structured Parzen Estimator (TPE)-Optimized eXtreme gradient boosting (XGBoost) for fine-grained prediction of tertiary vocational college students’ employment destinations. The dataset comprises 2,270 graduates (2016–2019 cohorts) from a single tertiary vocational college in China, characterized by 27 features spanning demographics, entrance scores, academic records, and extracurricular engagement. Employment destinations are categorized into three fine-grained classes: smooth employment, further study, and unemployment (class distribution: 78.8%, 17.8%, 3.4%). Recursive Feature Elimination (RFE) reduced the feature set to 22 informative variables. TPE hyperparameter optimization (100 iterations) was applied to the XGBoost classifier using an 80/20 train-test split, with stratified 5-fold cross-validation to mitigate overfitting risk. The proposed model achieves an overall accuracy of 81.72% and a balanced accuracy of 48.80% on the held-out test set, with macro-averaged precision of 0.6608, macro-averaged recall of 0.4880, macro-averaged specificity of 0.7659, and macro-averaged F1-score of 0.5263, outperforming seven baseline classifiers, including non-optimized XGBoost, logistic regression (LR), decision tree (DT), support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN), and LightGBM. SHAP analysis identifies academic GPA across multiple semesters, Gaokao scores, and volunteer service hours as the most influential predictors.

Related

  1. LKMHCpan: Transformer-optimized HLA-I pseudo-sequences enhance neoantigen prediction for immunotherapy
    Qing Li, Sanyang Liu, Mingyue Yao +5
    iSciencePeer-reviewedCC BY-NC-ND 4.0Bioinformatics & computational biologyImmunologyCrossrefCell Press
  2. An optimized machine learning approach for reliable agronomic parameter prediction in precision farming systems
    T. Suba, K. Lakshmi Joshitha
    Scientific ReportsPeer-reviewedCC BYFull textPlant biologyDOAJ
  3. NuFold RNA Structure Database: Large-Scale RNA Tertiary Structure Prediction with Confidence Metrics
    D. Platt, Y. Kagaya, D. Kihara
    bioRxivPreprintCC BY-NC-NDFull textBioinformatics & computational biologybioRxiv
  4. Resource integration spectrum of tourist destinations and its construction: A case study of Zunyi City
    XU Chunxiao, CHEN Jingjing, LIANG Huiting
    Ziyuan KexuePeer-reviewedCC BY-NC-NDFull textGeneral biologyEnvironmental science & pollutionDOAJ
  5. Stimulus over expertise: what governs clinical visual attention in early health-vocational students.
    Burcu Yuksel, Begüm Çapa Tayyare, Nurullah Eryilmaz
    Frontiers in medicinePeer-reviewedCC BYFull textNeuroscienceEurope PMCPMC

Cite

BibTeXRIS
BibTeX
@article{chen2026tpeoptimized,
  title = {{TPE-optimized XGBoost model for fine–grained prediction of tertiary vocational college students’ employment destinations}},
  author = {Juntao Chen and Jinmei Zhan and Xiaodeng Zhou and Jiahua Yao and Su-Kit Tang and Xia Liu},
  journal = {PLOS One},
  year = {2026},
  month = sep,
  doi = {10.1371/journal.pone.0359421},
  url = {https://doi.org/10.1371/journal.pone.0359421},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - TPE-optimized XGBoost model for fine–grained prediction of tertiary vocational college students’ employment destinations
AU  - Juntao Chen
AU  - Jinmei Zhan
AU  - Xiaodeng Zhou
AU  - Jiahua Yao
AU  - Su-Kit Tang
AU  - Xia Liu
PY  - 2026
DA  - 2026/09/28
JO  - PLOS One
DO  - 10.1371/journal.pone.0359421
UR  - https://doi.org/10.1371/journal.pone.0359421
AB  - Predicting tertiary vocational college students’ employment destinations has become an urgent need for most tertiary vocational colleges. This study proposes a machine learning framework based on Tree-structured Parzen Estimator (TPE)-Optimized eXtreme gradient boosting (XGBoost) for fine-grained prediction of tertiary vocational college students’ employment destinations. The dataset comprises 2,270 graduates (2016–2019 cohorts) from a single tertiary vocational college in China, characterized by 27 features spanning demographics, entrance scores, academic records, and extracurricular engagement. Employment destinations are categorized into three fine-grained classes: smooth employment, further study, and unemployment (class distribution: 78.8%, 17.8%, 3.4%). Recursive Feature Elimination (RFE) reduced the feature set to 22 informative variables. TPE hyperparameter optimization (100 iterations) was applied to the XGBoost classifier using an 80/20 train-test split, with stratified 5-fold cross-validation to mitigate overfitting risk. The proposed model achieves an overall accuracy of 81.72% and a balanced accuracy of 48.80% on the held-out test set, with macro-averaged precision of 0.6608, macro-averaged recall of 0.4880, macro-averaged specificity of 0.7659, and macro-averaged F1-score of 0.5263, outperforming seven baseline classifiers, including non-optimized XGBoost, logistic regression (LR), decision tree (DT), support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN), and LightGBM. SHAP analysis identifies academic GPA across multiple semesters, Gaokao scores, and volunteer service hours as the most influential predictors.
SN  - 1932-6203
ER  -