Skripsi

PENERAPAN REDUKSI DIMENSI DENGAN LINEAR DISCRIMINANT ANALYSIS PADA KLASIFIKASI PENYAKIT ARITMIA

---

Penilaian

---

Arrhythmia disease is a serious and potentially fatal condition, so detecting this disease early has great significance. However, the dataset in the form of electrocardiogram (ECG) used in the analysis of arrhythmia disease consists of attributes, namely 252 free attributes and 1 label. Thus, 252 signal points are taken which means 1 signals points are taken which means 1 signal has 252 attributes. The more attributes, the more memory is required, but computers have limited memory. To overcome this problem, dimensionality reduction is performed to reduce attributes without losing important information. Linear Discriminant Analysis (LDA) was chosen as one of the statistical feature extraction methods used to reduce the high dimensionality of data. This study aims to apply LDA in reducing the dimensionality of the MIT-BIH arrhythmia classification. The results of this attributes to 10 attributes and the Naive Bayes algorithm obtained an increase in accuracy of 21%, precision of 14%, recall of 21%, and f1-score of 28%. The KNN algorithm obtained an increase in accuracy of 3%, precision 2%, recall 3%, and f1-score 2%. The SVM algorithm obtained an increase in accuracy of 3%, precision of 3%, recall 4%, and f1-score 4%. The arrhythmia disease dataset experienced a significant increase ini precision, recall, and f1-score accuracy by 9%, 6%, 9%, and 11%, respectively. Based on the results ibtained, it can be concluded that dimension reduction with LDA improves classification performance on the MIT-BIH arrhythmia disease dataset.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2307006313	T130382	T1303822023	Central Library (Referens)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1303822023

Publisher

Inderalaya : Jurusan Matematika, Fakultas Matematika Dan Ilmu Pengetahuan Alam, Universitas Sriwijaya., 2023

Collation

xii, 65 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

510.285 07

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Jurusan Matematika
Pengolahan dan Analisa Data di Bidang Matematika

Specific Detail Info

-

Statement of Responsibility

SEPTA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment