Machine Learning for Stroke Prediction: Evaluating the Effectiveness of Data Balancing Approaches DOI Creative Commons

M Indra,

Siti Ernawati, Ilham Maulana

и другие.

Jurnal Riset Informatika, Год журнала: 2024, Номер 6(4), С. 211 - 222

Опубликована: Сен. 15, 2024

Stroke occurs due to disrupted blood flow the brain, either from a clot (ischemic) or ruptured vessel (hemorrhagic), leading brain tissue damage and neurological dysfunction. It remains cause of death disability worldwide, making early prediction crucial for timely intervention. This study evaluates impact data balancing techniques on stroke performance across different machine learning models. Random Forest (RF) consistently achieves highest accuracy (98%) but struggles with precision recall variations depending method. Decision Tree (DT) K-Nearest Neighbors (KNN) benefit most SMOTE SMOTETomek, improving their F1-scores (11.21% 9.18%), indicating better balance between recall. Under Sampling enhances all models reduces precision, lower overall predictive reliability. SMOTETomek emerge as effective techniques, particularly DT KNN, while RF accurate requires further optimization improve balance.

Язык: Английский

A predictive maintenance architecture for TFT-LCD manufacturing using machine learning on the cloud service DOI
Chih‐Hung Chang,

Hsin‐Ta Chiao,

Hsiang-Ching Chang

и другие.

Internet of Things, Год журнала: 2025, Номер unknown, С. 101541 - 101541

Опубликована: Фев. 1, 2025

Язык: Английский

Процитировано

0
The ensemble learning combined with the pruning model reveals the spectral response mechanism of tidal flat mapping in China DOI Creative Commons

Jiapeng Dong,

Kai Jia, Chongyang Wang

и другие.

Ecological Informatics, Год журнала: 2025, Номер unknown, С. 103104 - 103104

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Towards sustainable digital learning environments: unmasking the impact of media presentation on adolescents’ visual health DOI
Jiaqi Xu, A. Gauthier, Qian Liu

и другие.

Interactive Learning Environments, Год журнала: 2025, Номер unknown, С. 1 - 17

Опубликована: Апрель 15, 2025

Язык: Английский

Процитировано

0
Machine Learning for Stroke Prediction: Evaluating the Effectiveness of Data Balancing Approaches DOI Creative Commons

M Indra,

Siti Ernawati, Ilham Maulana

и другие.

Jurnal Riset Informatika, Год журнала: 2024, Номер 6(4), С. 211 - 222

Опубликована: Сен. 15, 2024

Stroke occurs due to disrupted blood flow the brain, either from a clot (ischemic) or ruptured vessel (hemorrhagic), leading brain tissue damage and neurological dysfunction. It remains cause of death disability worldwide, making early prediction crucial for timely intervention. This study evaluates impact data balancing techniques on stroke performance across different machine learning models. Random Forest (RF) consistently achieves highest accuracy (98%) but struggles with precision recall variations depending method. Decision Tree (DT) K-Nearest Neighbors (KNN) benefit most SMOTE SMOTETomek, improving their F1-scores (11.21% 9.18%), indicating better balance between recall. Under Sampling enhances all models reduces precision, lower overall predictive reliability. SMOTETomek emerge as effective techniques, particularly DT KNN, while RF accurate requires further optimization improve balance.

Язык: Английский

Процитировано

1