APPLICATION OF MULTISURF FOR FEATURE WEIGHTING AT THE RETRIEVAL STAGE OF CASE-BASED REASONING IN EARLY DETECTION OF TYPE 2 DIABETES MELLITUS

Ramadhani, Aimee Natya (2026) APPLICATION OF MULTISURF FOR FEATURE WEIGHTING AT THE RETRIEVAL STAGE OF CASE-BASED REASONING IN EARLY DETECTION OF TYPE 2 DIABETES MELLITUS. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 22081010159.-cover.pdf Download (2MB)
	Text (BAB 1) 22081010159.-bab1.pdf Download (121kB)
	Text (BAB 2) 22081010159.-bab2.pdf Restricted to Repository staff only until 2 September 2028. Download (1MB)
	Text (BAB 3) 22081010159.-bab3.pdf Restricted to Repository staff only until 2 September 2028. Download (2MB)
	Text (BAB 4) 22081010159.-bab4.pdf Restricted to Repository staff only until 2 September 2028. Download (3MB)
	Text (BAB 5) 22081010159.-bab5.pdf Download (109kB)
	Text (Daftar Pustaka) 22081010159.-daftarpustaka.pdf Download (94kB)
	Text 22081010159.-lampiran.pdf Restricted to Repository staff only until 2 September 2028. Download (480kB)

Abstract

Type 2 Diabetes Mellitus is a chronic disease that develops slowly without obvious early symptoms, so early detection is crucial to prevent further complications. This study aims to apply the Case-Based Reasoning (CBR) method with feature weighting using the MultiSURF algorithm and Weighted Euclidean Distance-based similarity measurement at the retrieval stage to support the early detection of Type 2 Diabetes Mellitus, using the Pima Indians Diabetes Dataset of 768 patient data with eight clinical features. The results showed that the Glucose feature obtained the highest weight followed by BMI and Age, with the best configuration at n_neighbors = 3 and K = 9 which resulted in an Accuracy of 76.03%, Precision 65.87%, Recall 64.54%, and an F1-Score of 64.95%. Compared to the CBR model without weighting (Baseline), the MultiSURF-based Weighted CBR model has succeeded in improving Accuracy, Precision, and F1-Score, so that this method can be a viable alternative in supporting the early detection of Type 2 Diabetes Mellitus based on clinical data. Keywords : Case-Based Reasoning, MultiSURF, Feature Weighting, Euclidean Distance, Type 2 Diabetes Mellitus

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Muttaqin, Faisal	NIDN0030058602	faisalmuttaqin.if@upnjatim.ac.id
Thesis advisor	Puspaningrum, Eva Yulia	NIDN0005078908	evapuspaningrum.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Aimee Natya Ramadhani

Date Deposited:

03 Sep 2026 06:55

Last Modified:

04 Sep 2026 03:42

URI:

https://repository.upnjatim.ac.id/id/eprint/59864