Skripsi

IMPUTASI DATA MEDIS HATI KRONIS MENGGUNAKAN ALGORITMA K- NEAREST NEIGHBORS IMPUTATION

---

Medical data related to chronic liver disease often contain missing values that can reduce the quality of data analysis and the accuracy of predictive models. Handling missing values properly is essential to ensure optimal classification performance. This study aims to apply and evaluate the K-Nearest Neighbor (KNN) Imputation method to address missing values in a medical dataset of liver cirrhosis patients and to analyze its impact on classification performance. The dataset used in this study was obtained from the UCI Machine Learning Repository and consists of 418 patient records divided into three classes: C (Censored/Recovered) with 258 records, D (Deceased) with 161 records, and CL (Transplanted) with 25 records. The dataset contains 1,069 missing values distributed across several predictor features, making data preprocessing a crucial step before classification. The preprocessing stages include data imputation using KNN Imputation, data normalization, and handling class imbalance using the Synthetic Minority Over sampling Technique (SMOTE). Experiments were conducted by varying the parameter k to analyze its effect on imputation quality and predictive model performance. The classification model employed in this study was the Support Vector Machine (SVM). The results show that the combination of KNN Imputation and SVM achieved the best performance, with an accuracy of 72.62% and an F1 score of 74.12% at k = 13. These findings demonstrate that KNN Imputation is effective in improving medical data quality and significantly enhances the performance of the SVM model in chronic liver disease classification. Keywords: KNN Imputation, Missing Value, Medical Data, Chronic Liver Disease, Machine Learning

---

Availability

Detail Information

Series Title

-

Call Number

T1901092025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xiii, 101 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

616.360 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Penyakit Hati--Nearest Neighbors Imputation

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment