Pratiwi, Julia Ningsi and Gasim, Gasim and Mair, Zaid Romegar (2026) Perbandingan Metode K-Nearest Neighbor dan Naive Bayes dalam Klasifikasi Tingkat Kesegaran Daging Ayam Ras Berdasarkan Fitur Warna. Masters thesis, Universitas Indo Global Mandiri.
Cover.pdf
Download (547kB) | Preview
Skripsi Full.pdf
Restricted to Repository staff only
Download (2MB) | Request a copy
2022110121_JUL_2026_1.pdf
Restricted to Repository staff only
Download (15MB) | Request a copy
Abstract
ABSTRAK
Kesegaran daging ayam ras menjadi salah satu aspek penting yang berhubungan dengan kualitas serta keamanan pangan. Proses penilaian kesegaran daging ayam pada umumnya masih dilakukan secara manual dengan memperhatikan karakteristik seperti warna, tekstur, dan aroma, sehingga penilaiannya berpotensi bersifat subjektif. Penelitian ini bertujuan untuk membandingkan performa metode K-NN dan Naïve Bayes dalam melakukan klasifikasi tingkat kesegaran daging ayam ras dengan memanfaatkan fitur warna RGB yang diperoleh dari citra digital. Dataset penelitian terdiri atas 300 citra yang dikelompokkan ke dalam tiga kelas, yaitu Segar, Kurang Segar, dan Busuk, dengan jumlah masing-masing kelas sebanyak 100 citra. Tahapan penelitian mencakup pra-pemrosesan citra, ekstraksi fitur warna RGB, normalisasi data, pembagian data latih dan data uji dengan rasio 80:20, klasifikasi menggunakan metode K-NN dan Naïve Bayes, serta evaluasi kinerja menggunakan confusion matrix. Pada penerapan metode K-NN, dilakukan pengujian terhadap nilai K sebesar 1, 3, 5, 7, dan 9 untuk memperoleh nilai K yang memberikan hasil klasifikasi paling optimal. Hasil penelitian menunjukkan bahwa metode K-NN dengan K=5 memperoleh accuracy sebesar 66,67%, precision sebesar 0,683, recall sebesar 0,667, dan F1-score sebesar 0,671. Sementara itu, metode Naïve Bayes memperoleh accuracy sebesar 63,33%, precision sebesar 0,624, recall sebesar 0,633, dan F1-score sebesar 0,628. Berdasarkan hasil evaluasi tersebut, metode K-NN memiliki performa yang lebih baik dibandingkan dengan metode Naïve Bayes dalam melakukan klasifikasi tingkat kesegaran daging ayam ras berdasarkan fitur warna RGB.
Kata Kunci: Kesegaran daging ayam, K-Nearest Neighbor, Naïve Bayes, RGB, klasifikasi citra.
ABSTRACT
The freshness of broiler chicken meat is one of the important factors determining food quality and safety. The assessment of chicken meat freshness is generally still carried out manually based on characteristics such as color, texture, and aroma, making the assessment potentially subjective. This study aims to compare the performance of the K-Nearest Neighbor (K-NN) and Naïve Bayes methods in classifying the freshness level of broiler chicken meat based on RGB color features extracted from digital images. The dataset consisted of 300 images divided into three classes, namely Fresh, Less Fresh, and Rotten, with 100 images in each class. The research stages included image preprocessing, RGB color feature extraction, data normalization, division of training and testing data using an 80:20 ratio, classification using the K-NN and Naïve Bayes methods, and performance evaluation using a confusion matrix. In the K-NN method, several K values, namely 1, 3, 5, 7, and 9, were tested to determine the K value that provided the most optimal classification results. The results showed that the K-NN method with K=5 achieved an accuracy of 66.67%, precision of 0.683, recall of 0.667, and F1-score of 0.671. Meanwhile, the Naïve Bayes method achieved an accuracy of 63.33%, precision of 0.624, recall of 0.633, and F1-score of 0.628. Based on the evaluation results, the K-NN method demonstrated better performance than the Naïve Bayes method in classifying the freshness level of broiler chicken meat based on RGB color features.
Keywords: Broiler chicken freshness, K-Nearest Neighbor, Naïve Bayes, RGB, image classification.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer dan Sains > Teknik Informatika S1 |
| Depositing User: | Julia Ningsi Pratiwi Herlambang |
| Date Deposited: | 10 Aug 2026 09:12 |
| Last Modified: | 10 Aug 2026 09:12 |
| URI: | https://repository.uigm.ac.id/id/eprint/6944 |
