Mahendra, Tatia (2026) Prediksi Risiko Diabetes Menggunakan Multi-Layer Perceptron Dengan Analisis Feature Importance. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5024211073_Undergraduate Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5024211073_Undergraduate Thesis.pdf Restricted to Repository staff only Download (4MB) | Request a copy |
Abstract
Diabetes merupakan salah satu penyakit kronis yang dikenal silent killer dikarenakan banyak penderita baru menyadari penyakit diabetes ketika sudah berkembang ke kondisi lebih serius. Oleh sebab itu, deteksi dini merupakan salah satu langkah penting untuk mengetahui adanya diabetes. Penelitian ini berfokus pada pengembangan model gabungan Machine Learning dan Deep Learning berbasis Multi-Layer Perceptron untuk memprediksi risiko diabetes. Dari pengembangan model, ada hasil evaluasi berupa performa klasifikasi untuk menitikberatkan pada analisa interpretabilitas model melalui Feature Importance menggunakan skala global dan skala lokal. Pada skala global, menggunakan Permutation Feature Importance untuk identifikasi fitur-fitur apa saja yang paling berpengaruh dari keseluruhan dataset. Sedangkan, pada skala lokal, digunakan metode SHAP (SHapley Additive exPlanations) untuk memberikan pemahaman mengenai kontribusi pada setiap fitur yang muncul berdasarkan hasil prediksi pada individual. Sebagai bentuk visualisasi hasil penelitian, model yang dikembangkan diintegrasikan ke dalam sebuah website berbasis Streamlit yang menyediakan informasi terkait prediksi risiko diabetes serta fitur-fitur yang mempengaruhi risiko diabetes. Hasil dari penelitian ini diharapkan dapat memberikan kontribusi dalam pengembangan model prediksi diabetes yang memiliki pemahaman yang dapat dipahami dengan mudah sehingga dapat membantu pemahaman terhadap faktor-faktor yang berkontribusi terhadap risiko diabetes serta mendukung pengambilan keputusan terutama di bidang kesehatan.
==================================================================================================================================
Diabetes is a chronic disease commonly referred to as a silent killer, as many patients only become aware of their condition when it has progressed to a more severe stage. Therefore, early detection is an essential step in identifying the presence of diabetes. This study focuses on the development of a combined Machine Learning and Deep Learning model based on a Multi-Layer Perceptron (MLP) to predict diabetes risk. In addition to evaluating classification performance, this research emphasizes model interpretability analysis through Featue Importance at both global and local scales. At the global scale, Permutation Feature Importance is applied to identify the most influential features across the entire dataset. Meanwhile, at the local scale, the SHAP (SHapley Additive exPlanations) method is employed to provide insights into the contribution of each feature to individual prediction outcomes. As a form of result visualization,] the developed model is integrated into a Streamlit-based website that presents diabetes risk predictions along with explanations of the features influencing the risk. The results of this study are expected to contribute to the development of diabetes prediction models that are not only accurate but also interpretable, thereby enhancing the understanding of factors contributing to diabetes risk and supporting decision-making processes, particularly in the healthcare domain.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Diabetes Melitus, Prediksi Risiko, Multi-Layer Perceptron, Feature Importance, Permutation Feature Importance, SHAP Diabetes Mellitus, Risk Prediction, Multi-Layer Perceptron, Streamlit, Feature Importance, Permutation Feature Importance, SHAP |
| Subjects: | Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. R Medicine > R Medicine (General) > R858 Deep Learning |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis |
| Depositing User: | Tatia Mahendra |
| Date Deposited: | 21 Jan 2026 08:54 |
| Last Modified: | 21 Jan 2026 08:54 |
| URI: | http://repository.its.ac.id/id/eprint/130005 |
Actions (login required)
 |
View Item |