Skripsi

KLASIFIKASI DATA PENYAKIT DIABETES MENGGUNAKAN ALGORITMA DECISION TREE DAN PARTICLE SWARM OPTIMIZATION

---

The prevalence of diabetes is very high in Indonesia so diabetes classification is crucial for early diagnosis in detecting this disease. The classification of diabetes data carried out in this study used an algorithm Decision Tree C4.5. However, the algorithm decision tree has weaknesses when handling large datasets, where not all features are relevant for the classification process, which can reduce the level of accuracy. One of the algorithms that can be integrated with the C4.5 decision tree algorithm to select relevant features is Particle Swarm Optimization (PSO). This research also uses a data balancing method to obtain more accurate results. The data balancing method used is Synthetic Minority Over-sampling Technique (SMOTE) day Random Undersampling (RUS). The research results show that the combination of the C4.5 algorithm with PSO in feature selection and the data balancing method significantly increases the accuracy of diabetes classification, producing the highest accuracy of 83.33% and an increase in accuracy of 16.33% compared to using C4.5 without PSO. The optimal PSO parameters are number of particles=10, C1 value=2, C2 value=2, and maximum iteration=10.

---

Availability

Detail Information

Series Title

-

Call Number

T1558672024

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xiii, VI-1 hlm.; ilus.; tab, 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

unmediated

Carrier Type

other (computer)

Edition

-

Subject(s)

Kecerdasan Buatan
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

SEW

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment