EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

An An Optimized Hybrid Machine Learning Framework for Parkinson’s disease Mortality Prediction using Improved Harris Hawks Optimization and CatBoost

KARTHIGA HARIKRISHNA, Sumitra P

International Journal of Computer Sciences and Engineering · 31.03.2026 · 14 (3) · 50-58

DOI: 10.26438/ijcse.v14i3.7282 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Computer Sciences and Engineering, online erschienen am 31.03.2026. Verfasst von KARTHIGA HARIKRISHNA und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Predicting mortality based on Parkinson’s disease (PD) severity is vital for effective clinical intervention and personalized care. However, high-dimensional biomedical data often contain redundant features that limit prediction accuracy and model generalizability. This study proposes a robust computational framework for PD mortality prediction by integrating advanced feature selection with optimized gradient-boosting classifiers. A hybrid model is developed using a Butterfly-Optimized Genetic Algorithm (BOGA) for optimal feature selection and a CatBoost classifier for mortality prediction. To further enhance exploration and exploitation capabilities, a Harris Hawks Optimization-based CatBoost model is introduced, along with an Improved Harris Hawks Optimization (IHHO) algorithm to overcome premature convergence and limited search efficiency. The Parkinson's Disease dataset from the UCI Machine Learning Repository is used to assess the suggested models, and they are contrasted with both conventional and cutting-edge ensemble machine learning classifiers. The IHHO-CatBoost model regularly beats baseline techniques in key performance parameters, including as accuracy, precision, recall, and F1-score, according to experimental data. The proposed framework offers an efficient and scalable clinical decision support tool for early PD severity assessment and personalized treatment planning.

Zitieren

APA

HARIKRISHNA, K., & P, S. (2026). An An Optimized Hybrid Machine Learning Framework for Parkinson’s disease Mortality Prediction using Improved Harris Hawks Optimization and CatBoost. International Journal of Computer Sciences and Engineering, 14(3), 50-58. https://doi.org/10.26438/ijcse.v14i3.7282

Vancouver

HARIKRISHNA K, P S. An An Optimized Hybrid Machine Learning Framework for Parkinson’s disease Mortality Prediction using Improved Harris Hawks Optimization and CatBoost. International Journal of Computer Sciences and Engineering. 2026;14(3):50-58. doi:10.26438/ijcse.v14i3.7282

BibTeX
@article{HARIKRISHNA2026,
  author = {HARIKRISHNA, KARTHIGA and P, Sumitra},
  title = {An An Optimized Hybrid Machine Learning Framework for Parkinson’s disease Mortality Prediction using Improved Harris Hawks Optimization and CatBoost},
  journal = {International Journal of Computer Sciences and Engineering},
  year = {2026},
  volume = {14},
  number = {3},
  pages = {50--58},
  doi = {10.26438/ijcse.v14i3.7282}
}
Kostenlos starten →