Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods
Hypertension is a primary factor in diseases such as stroke, heart failure, myocardial infarction, atrial fibrillation, peripheral arterial disease, and aortic dissection. Early detection of hypertension from medical history is very urgent for the first treatment of patients so that the patient'...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IJMADA
2023-03-01
|
| Series: | International Journal of Management and Data Analytics |
| Subjects: | |
| Online Access: | https://ijmada.com/index.php/ijmada/article/view/14 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832593417641656320 |
|---|---|
| author | M. Hafidz Ariansyah Sri Winarno |
| author_facet | M. Hafidz Ariansyah Sri Winarno |
| author_sort | M. Hafidz Ariansyah |
| collection | DOAJ |
| description | Hypertension is a primary factor in diseases such as stroke, heart failure, myocardial infarction, atrial fibrillation, peripheral arterial disease, and aortic dissection. Early detection of hypertension from medical history is very urgent for the first treatment of patients so that the patient's life expectancy increases, increases the effectiveness of treatment, reduces treatment costs, and reduces the severity of hypertension. Researchers get detection results using a branch of AI technology, namely machine learning to find new knowledge from data and find patterns to make diagnoses. Researchers use machine learning that can explore large amounts of data sets to produce knowledge that is beneficial to science. In this paper, the researchers used the Passive-Aggressive algorithm and the PA-I and PA-II methods to make a model for the diagnosis of hypertension. This algorithm can work well for learning by transforming data and dealing with unbalanced classification problems. PA-I shows stable accuracy of test data with a value of 80.3 - 84.15%, and PA-II shows accuracy instability with a value of 71.41 - 82.41%. From these results, PA-I shows that the model is good in diagnosing hypertension patients because its accuracy is stable and high enough. The results also show that the model is not overfitting, and the new data can be predicted well in line with the training data because, on the results of training accuracy, PA-I shows an accuracy of 81.6 - 84.56% while PA-II shows an accuracy of 71.6 - 82.71%. |
| format | Article |
| id | doaj-art-723fcaec684f49c3b7ed854183d1c0a2 |
| institution | Kabale University |
| issn | 2816-9395 |
| language | English |
| publishDate | 2023-03-01 |
| publisher | IJMADA |
| record_format | Article |
| series | International Journal of Management and Data Analytics |
| spelling | doaj-art-723fcaec684f49c3b7ed854183d1c0a22025-01-20T15:45:31ZengIJMADAInternational Journal of Management and Data Analytics2816-93952023-03-011610.5281/zenodo.1152757014Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II MethodsM. Hafidz Ariansyah0Sri Winarno1Dian Nuswantoro UniversityDian Nuswantoro UniversityHypertension is a primary factor in diseases such as stroke, heart failure, myocardial infarction, atrial fibrillation, peripheral arterial disease, and aortic dissection. Early detection of hypertension from medical history is very urgent for the first treatment of patients so that the patient's life expectancy increases, increases the effectiveness of treatment, reduces treatment costs, and reduces the severity of hypertension. Researchers get detection results using a branch of AI technology, namely machine learning to find new knowledge from data and find patterns to make diagnoses. Researchers use machine learning that can explore large amounts of data sets to produce knowledge that is beneficial to science. In this paper, the researchers used the Passive-Aggressive algorithm and the PA-I and PA-II methods to make a model for the diagnosis of hypertension. This algorithm can work well for learning by transforming data and dealing with unbalanced classification problems. PA-I shows stable accuracy of test data with a value of 80.3 - 84.15%, and PA-II shows accuracy instability with a value of 71.41 - 82.41%. From these results, PA-I shows that the model is good in diagnosing hypertension patients because its accuracy is stable and high enough. The results also show that the model is not overfitting, and the new data can be predicted well in line with the training data because, on the results of training accuracy, PA-I shows an accuracy of 81.6 - 84.56% while PA-II shows an accuracy of 71.6 - 82.71%.https://ijmada.com/index.php/ijmada/article/view/14diagnosticclassificationhypertensionmodelpassiveagrresive |
| spellingShingle | M. Hafidz Ariansyah Sri Winarno Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods International Journal of Management and Data Analytics diagnostic classification hypertension model passiveagrresive |
| title | Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods |
| title_full | Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods |
| title_fullStr | Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods |
| title_full_unstemmed | Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods |
| title_short | Hypertension Detection Using Passive-Aggressive Algorithm With The PA-I And PA-II Methods |
| title_sort | hypertension detection using passive aggressive algorithm with the pa i and pa ii methods |
| topic | diagnostic classification hypertension model passiveagrresive |
| url | https://ijmada.com/index.php/ijmada/article/view/14 |
| work_keys_str_mv | AT mhafidzariansyah hypertensiondetectionusingpassiveaggressivealgorithmwiththepaiandpaiimethods AT sriwinarno hypertensiondetectionusingpassiveaggressivealgorithmwiththepaiandpaiimethods |