Yuandhika, Adelia (2026) Klasifikasi Subtipe Anemia Menggunakan Metode Extremely Randomized Trees dengan Optimalisasi Hyperparameter Accelerated Particle Swarm Optimization (APSO). Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
coverr skripsi final.pdf Download (1MB) | Preview |
|
|
Text (Bab 1)
22082010066 - bab1.pdf Download (300kB) | Preview |
|
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
22083010066 - bab2.pdf Restricted to Repository staff only until 7 July 2029. Download (618kB) |
|
|
Text (Bab 3)
22083010066 - bab3.pdf Restricted to Repository staff only until 7 July 2029. Download (433kB) |
|
|
Text (Bab 4)
22083010066 - bab4.pdf Restricted to Repository staff only until 7 July 2029. Download (2MB) |
|
|
Text (Bab 5)
22083010066 - bab5.pdf Download (217kB) | Preview |
|
|
Text (Daftar Pustaka)
22083010066 - daftarpustaka.pdf Download (180kB) | Preview |
|
|
Text (Lampiran)
22083010066 - lampiran.pdf Restricted to Repository staff only until 7 July 2029. Download (723kB) |
Abstract
Anemia is a health condition that negatively affects both physical well-being and social functioning. Given its moderate-to-high prevalence among the Indonesian population, effective early identification methods are essential to reduce its incidence. This study employs a machine learning approach to automatically classify anemia subtypes, thereby supporting targeted intervention decisions and follow-up examinations. The Extremely Randomized Trees algorithm was selected for its hierarchical decision tree structure combined with high randomization, which confers resistance to overfitting and strong generalization capability. To further enhance model performance, Accelerated Particle Swarm Optimization was applied as the hyperparameter optimization method, tuning five parameters, n_estimators, max_depth, min_samples_split, min_samples_leaf, and max_features. The dataset was obtained from Haji Provincial General Hospital in East Java and comprises three classes: iron-deficiency anemia, anemia of chronic disease, and non-anemia. Input variables included patient demographic data alongside complete blood count results, encompassing Hb, HCT, RBC, MCV, MCH, MCHC, and RDW-CV. Evaluation using a 90:10 holdout split demonstrated that the optimized model achieved strong classification performance, with an accuracy of 0.97, precision of 0.96, recall of 0.97, and an F1-score of 0.97.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics Q Science > QA Mathematics > QA76.6 Computer Programming |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Data Science | ||||||||||||
| Depositing User: | Adelia Yuandhika | ||||||||||||
| Date Deposited: | 07 Jul 2026 07:25 | ||||||||||||
| Last Modified: | 07 Jul 2026 07:25 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/54730 |
Actions (login required)
 |
View Item |