Skripsi

OPTIMASI ALGORITMA K-NEAREST NEIGHBOR (KNN) DENGAN NORMALISASI DAN SELEKSI FITUR UNTUK KLASIFIKASI PENYAKIT KOMPLIKASI INFARK MIOKARD

---

Myocardial infarction complications require early, accurate prediction for clinical intervention. This study optimizes the K-nearest neighbor (KNN) algorithm to classify these complications using Z-Score normalization and three feature selection methods: Information Gain, Gain Ratio, and Symmetrical Uncertainty.Using a dataset from the UCI Machine Learning Repository, preprocessing included mean imputation for missing values and the Synthetic Minority Over-sampling Technique (SMOTE) to resolve class imbalance. The model was tested across various K values (3, 5, 7, 9) on both raw and normalized data.Results demonstrated that Z-score normalization consistently improved overall accuracy by 2% to 3%. Information Gain achieved the highest accuracy of 93% at K=3 with normalized data. Symmetrical Uncertainty yielded a maximum accuracy of 77%, while Gain Ratio performed poorest at 46%. In conclusion, integrating Z-score normalization with Information Gain significantly enhances the KNN algorithm's accuracy in classifying myocardial infarction complications.

---

Availability

Detail Information

Series Title

-

Call Number

T2012632026

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xiii, 92 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

006.37

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Komputer
Algoritma--Data Mining

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment