Publisher

PDSIPDSI

Journal Bulletin of Informatics and Data ScienceBulletin of Informatics and Data Science Abstract

Diabetes is a major cause of many chronic diseases such as visual impairment, stroke and kidney failure. Early detection especially in groups that have a high risk of developing diabetes needs to be done to prevent problems that have a wide impact. Indonesia is ranked seventh in the world with a prevalence of 10.7% of the total number of people with diabetes. This research aims to determine the attributes in the diabetes dataset that most affect the classification and apply the Support Vector Machine method for diabetes classification. For the determination process, Gain Ratio feature selection technique is applied. The dataset used consists of 768 data with 8 attributes. In this classification process, 3 SVM kernels (Linear, Polynomial, and RBF) are used with three possible data divisions using the ratio (70:30; 80:20; 90:10). Before applying feature selection, there were 8 attributes used and achieved the highest accuracy of 94.81% at a ratio of 80:20 using the RBF kernel with a combination of two parameters namely C = 100, Gamma = 3 and C = 100, Gamma = Scale. Feature selection parameters in the form of thresholds used include 0.02; 0.03; and 0.05. After applying feature selection, the attribute that produces the highest accuracy uses 6 attributes. The highest accuracy after applying feature selection reached 95.45% at a threshold of 0.02 with a ratio of 80:20 using the RBF kernel with parameters C = 100 and Gamma = Scale. The results showed that there was an increase in accuracy after applying feature selection.

Conclusion

The application of the Gain Ratio feature selection in the Support Vector Machine method successfully increased the accuracy of diabetes classification by 0.20 data ratio, and the RBF kernel with parameters C = 100 and Gamma = Scale.The RBF kernel consistently demonstrated optimal performance.The right combination of threshold, data ratio, and kernel parameters can produce a more reliable model for predicting diabetes risk.

Future Research

Penelitian lebih lanjut dapat dilakukan dengan mengeksplorasi teknik data balancing untuk mengatasi ketidakseimbangan jumlah data antara kelas diabetes dan non-diabetes, karena hal ini berpotensi meningkatkan kinerja model klasifikasi. Selain itu, penelitian dapat diperluas dengan menguji kombinasi parameter yang lebih beragam pada algoritma SVM, termasuk penggunaan teknik optimasi parameter otomatis untuk menemukan konfigurasi terbaik. Sebagai pengembangan, studi komparatif dapat dilakukan dengan menerapkan teknik seleksi fitur Gain Ratio pada algoritma machine learning lainnya, seperti Random Forest atau Decision Tree, untuk mengidentifikasi algoritma mana yang paling efektif dalam memprediksi risiko diabetes. Penelitian ini juga dapat diperkaya dengan mempertimbangkan faktor-faktor lain yang mungkin mempengaruhi risiko diabetes, seperti data gaya hidup, riwayat keluarga, atau informasi genetik, untuk membangun model prediksi yang lebih komprehensif dan akurat. Terakhir, validasi model pada dataset yang lebih besar dan beragam dari berbagai populasi dapat dilakukan untuk memastikan generalisasi dan keandalan model dalam aplikasi dunia nyata.

References

1. Komparasi Information Gain, Gain Ratio, CFs-Bestfirst dan CFs-PSO Search Terhadap Performa Deteksi Anomali... ejurnal.stmik-budidarma.ac.id/index.php/mib/article/view/2258Komparasi Information Gain Gain Ratio CFs Bestfirst dan CFs PSO Search Terhadap Performa Deteksi Anomali ejurnal stmik budidarma ac index php mib article view 2258
2. Performance Analysis of LVQ 1 Using Feature Selection Gain Ratio for Sex Classification in Forensic Anthropology... ejurnal.seminar-id.com/index.php/bits/article/view/3625Performance Analysis of LVQ 1 Using Feature Selection Gain Ratio for Sex Classification in Forensic Anthropology ejurnal seminar id index php bits article view 3625

Metadata

File size	706.23 KB
Pages	12
DMCA	Report