EUVIMEDEUVIMED entdecken
Crossref

HYBRID MACHINE LEARNING MODELS FOR ACCURATE LIVER DISEASE PREDICTION

CHAVAN SHARAF NANDITH SAI, A RAVI SANKAR

ETDT · 26.03.2026 · 276

DOI: 10.65713/etdtv2i129 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ETDT, online erschienen am 26.03.2026. Verfasst von CHAVAN SHARAF NANDITH SAI und 1 weiteren Autor:innen. Die Arbeit nennt 10 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Early liver disease prediction helps clinicians treat patients faster and better. Clinical datasets with demographic, biochemical, and lifestyle data are prepared using various ways. This includes outlier treatment, missing value estimate, and normalization. Selecting the right features improves model performance and noise reduction. A hybrid machine learning model uses KNN, Random Forest, Support Vector Machines, and Deep Neural Networks to make more accurate predictions. Ensemble techniques aim to reduce prediction errors and optimize classifier performance. This approach distinguishes healthy people from liver disease patients. Model effectiveness is measured by F1-score, confusion matrix, recall, accuracy, precision, and recall. Testing suggests they outperform situational models. Some patient populations and data distributions benefit from the hybrid approach. Cross-validation ensures reliable and fair performance estimates. When we consider feature value, finding crucial clinical markers that affect predictions is easy.

Zitieren

APA

SAI, C. S. N., & SANKAR, A. R. (2026). HYBRID MACHINE LEARNING MODELS FOR ACCURATE LIVER DISEASE PREDICTION. ETDT, 276. https://doi.org/10.65713/etdtv2i129

Vancouver

SAI CSN, SANKAR AR. HYBRID MACHINE LEARNING MODELS FOR ACCURATE LIVER DISEASE PREDICTION. ETDT. 2026:276. doi:10.65713/etdtv2i129

BibTeX
@article{SAI2026,
  author = {SAI, CHAVAN SHARAF NANDITH and SANKAR, A RAVI},
  title = {HYBRID MACHINE LEARNING MODELS FOR ACCURATE LIVER DISEASE PREDICTION},
  journal = {ETDT},
  year = {2026},
  pages = {276},
  doi = {10.65713/etdtv2i129}
}
Kostenlos starten →