A STROKE DISEASE PREDICTION ANALYSIS USING LINEAR REGRESSION ALGORITHM ON PATIENT DATA
Article Sidebar
Main Article Content
Abstract
Stroke is a major cause of death and disability worldwide , including in Indonesia. Key risk factors include age , height blood glucose , hypertension , and unhealthy lifestyles . Early identification of stroke risk is essential for prevention . This study applies a linear regression algorithm to predict stroke based on patient data, focusing on age and average glucose level. The dataset , obtained from Kaggle , consists of 5,110 patients records with 11 health attributes . Research stages included data cleaning , selection , and transformation before modeling with linear regression . Validation used the Split Validation method (90% training and 10% testing), while model performance was measured with Root Mean Squared Error (RMSE). Results show that linear regression predicts stroke status with an RMSE of 0.218 ± 0.000, indicating low prediction error analysis confirms that increasing age and glucose levels significantly raise stroke risk . In conclusion , linear regression is a simple yet effective method for early stroke risk identification . This model may serve as the basis for a decision support system , enabling healthcare professionals and the public to improve preventive measures
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
[1] A. Puspitawuri, E. Santoso, and C. Dewi, “Diagnosis of Stroke Risk Level Using the K-Nearest Neighbor and Naïve Bayes Methods,” vol. 3, no. 4, pp. 3319–3324, 2019.
[2] A. Byna and M. Basit, “Application of Adaboost Method to Optimize Stroke Disease Prediction with Naïve Bayes Algorithm,” J. Sisfokom (Inf. and Computer Systems) , vol. 9, no. 3, pp. 407–411, 2020, doi: 10.32736/sisfokom.v9i3.1023.
[3] J. Antares, T. Infomasi, and U. Dharmawangsa, “Artificial Neural Network in Identifying Stroke Disease Using Backpropagation Method (Case Study at Madya Padang Pharmacy Clinic),” vol. 1, no. 1, pp. 6–14, 2020.
[4] ES Renaldy Senna Hutama, Nurul Hidayat, “Expert System for Early Detection of Stroke Using the Naive Bayes-Certainty Factor Method,” J. Pengemb. Teknol. Inf. dan Ilmu Comput. , vol. 2, no. 11, pp. 4333–4339, 2018.
[5] NO Rahardiani, “Optimization Of Multi-Layer Perceptron Weights Using Genetic Algorithm For Classification Of Stroke Risk Level, obtained a Bachelor of Computer Science degree,” 2017.
[6] R. Rachman et al. , “Implementation of Case Based Reasoning to Diagnose Stroke Using Probabilistic Symmetric Algorithm,” vol. 8, no. 1, pp. 10–16, 2021.
[7] K. Algoritma, C. Berbasis, PSO Dan, RS Rohman, RA Saputra, and DA Firmansaha, “Comparison of pso and ga based c4.5 algorithms for stroke diagnosis,” vol. 5, no. 1, pp. 155–161, 2020.
[8] DW Nugraha, AYE Dodu, and N. Chandra, “Stroke Disease Classification Using the Naive Bayes Classifier Method (Case Study at Undata Regional General Hospital, Palu),” semanTIK , vol. 3, no. 2, pp. 13–22, 2017.
[9] SAJI Nugroho, “Comparative publication manuscript of fuzzy k-nearest neighbor and weighted k-nearest neighbor methods for stroke detection,” 2020.
[10] D. Aprilla, C, B. Donny, Aji, A. Lia, and W. I, Wayan, Simri, “DATA MINING with RAPID MINER,” The large number of free and open source software products has made it easier for us to carry out data processing and analysis. In carrying out data mining analysis, RapidMiner is one of the , pp. 1–128, 2013.
[11] R. Fernando, L. Anggraini, and A. Nazir, “Analysis of the Relationship between Stroke Risk Factors and Type of Stroke Suffered Using the ECLAT Algorithm,” vol. 9789, pp. 18–19, 2017.
[12] Simple ARL, “Regression analysis,” vol. 0, pp. 29–52.
[13] R. Mean et al. , “Evaluation and Validation of Evaluation.”