AZIZI, SAID RAHMAN (2025) ANALISIS KLASIFIKASI KUALITAS KABEL TEGANGAN MENENGAH ISOLASI XLPE DENGAN ALGORITMA K - NEAREST NEIGHBOR. S1 thesis, Fakultas Teknik Universitas Sultan Ageng Tirtayasa.

	Text (Fulltext) Said Rahman Azizi_3332210079_Fulltext.pdf Restricted to Registered users only Download (3MB)
	Text (Bab 1) Said Rahman Azizi_3332210079_01.pdf Restricted to Registered users only Download (833kB)
	Text (Bab 2) Said Rahman Azizi_3332210079_02.pdf Restricted to Registered users only Download (609kB)
	Text (Bab 3) Said Rahman Azizi_3332210079_03.pdf Restricted to Registered users only Download (361kB)
	Text (Bab 4) Said Rahman Azizi_3332210079_04.pdf Restricted to Registered users only Download (776kB)
	Text (Bab 5) Said Rahman Azizi_3332210079_05.pdf Restricted to Registered users only Download (7kB)
	Text (Daftar Referensi) Said Rahman Azizi_3332210079_Ref.pdf Restricted to Registered users only Download (141kB)
	Text (Lampiran) Said Rahman Azizi_3332210079_Lamp.pdf Restricted to Registered users only Download (1MB)
	Text (Cek Plagiasi) Said Rahman Azizi_3332210079_CP.pdf Restricted to Registered users only Download (3MB)

Abstract

Over time, the quality of the cable decreases. This study conducted a cable assessment test and the results were classified using the K-Nearest Neighbor algorithm to make it easier to determine quality. The cable conditions were divided into 5 classes and the evaluation was carried out by testing the model performance using accuracy, precision, recall, and F1-score. The test results show that increasing the amount of training data can improve classification performance. With the addition of 100 training data, the K-NN algorithm with K=2 achieved the highest accuracy of 92.86%, precision 0.96, recall 0.9333, and F1-score 0.9378. These results indicate that K-NN can be used to classify underground cables, if supported by sufficient data and the selection of the right K parameters.

Actions (login required)

View Item