EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

PotentialNet: Explainable Employee Potential Prediction via Multi-View Representation Learning

Elena Park, Marcus Voss

International Journal of Advanced Engineering and Technology Research · 17.07.2026 · 2 (3) · 35-41

DOI: 10.54097/bgmh8n87 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Advanced Engineering and Technology Research, online erschienen am 17.07.2026. Verfasst von Elena Park und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Employee potential assessment is a high-impact human resource analytics problem, but it differs from promotion prediction because potential is prospective, multi-dimensional, and rarely available as a clean public label. This paper proposes PotentialNet, an explainable multi-view representation learning framework that separates employee attributes into profile, performance, growth, and behavior views. Dedicated view encoders learn view-specific representations, a contrastive alignment objective encourages agreement across complementary views of the same employee, and an attention fusion module produces both a high-potential probability and a view-level explanation. Because public data with verified potential labels was not available, we construct a fully reproducible semi-synthetic HR benchmark rather than inventing real employee data. Experiments compare PotentialNet with logistic regression, random forest, gradient boosting, multilayer perceptron, and LightGBM. Results show that PotentialNet is competitive with strong tabular baselines while providing interpretable view-level explanations and counterfactual action audits. The study is deliberately conservative: it avoids claims of real-world deployment validity and releases all code, generated data, figures, and raw results to support transparent reproduction.

Zitieren

APA

Park, E., & Voss, M. (2026). PotentialNet: Explainable Employee Potential Prediction via Multi-View Representation Learning. International Journal of Advanced Engineering and Technology Research, 2(3), 35-41. https://doi.org/10.54097/bgmh8n87

Vancouver

Park E, Voss M. PotentialNet: Explainable Employee Potential Prediction via Multi-View Representation Learning. International Journal of Advanced Engineering and Technology Research. 2026;2(3):35-41. doi:10.54097/bgmh8n87

BibTeX
@article{Park2026,
  author = {Park, Elena and Voss, Marcus},
  title = {PotentialNet: Explainable Employee Potential Prediction via Multi-View Representation Learning},
  journal = {International Journal of Advanced Engineering and Technology Research},
  year = {2026},
  volume = {2},
  number = {3},
  pages = {35--41},
  doi = {10.54097/bgmh8n87}
}
Kostenlos starten →