Arif, Hidayat and Sutedi, Sutedi (2025) KLASIFIKASI RAMBUT RONTOK MENGGUNAKAN METODE NAIVE BAYES. Masters thesis, Institut Informatika dan Bisnis Darmajaya.
![[img]](http://repo.darmajaya.ac.id/style/images/fileicons/text.png) |
Text
1. Cover.pdf Download (685kB) |
|
Text
Halaman Pernyataan.pdf Download (41kB) |
|
Text
2. Abstrak.pdf Download (190kB) |
|
Text
3. Halaman Pengesahan.pdf Download (221kB) |
|
Text
4. Daftar Isi.pdf Download (687kB) |
|
Text
5. BAB 1.pdf Download (685kB) |
|
Text
6. BAB 2.pdf Download (685kB) |
|
Text
7. BAB 3.pdf Download (685kB) |
|
Text
8. BAB 4.pdf Download (685kB) |
|
Text
9. BAB 5.pdf Download (685kB) |
|
Text
10. Daftar Pustaka.pdf Download (685kB) |
|
Text
11. Lampiran.pdf Download (76kB) |
Abstract
INTISARI KLASIFIKASI RAMBUT RONTOK MENGGUNAKAN METODE NAÏVE BAYES Oleh Arif Hidayat arifhidayat124@gmail.com Rambut merupakan bagian tubuh yang berperan penting dalam penampilan dan kesehatan, sehingga kondisinya kerap menjadi perhatian khusus. Rambut rontok adalah salah satu masalah yang paling umum dialami masyarakat, dengan penyebab beragam seperti faktor genetik dan gaya hidup. Berdasarkan survei Lembaga Jajak Pendapat (Jakpat) pada pertengahan 2023 terhadap 3.041 responden, masalah rambut paling banyak adalah rambut rontok (64,7%), diikuti ketombe (44,3%), rambut kering dan kusam (30,8%), rambut berminyak (26,1%), dan rambut bercabang (18%). Mayoritas kasus rambut rontok dialami kelompok usia 20–25 tahun (37,7%). Permasalahan ini mendorong upaya deteksi dini dan perawatan yang tepat untuk menjaga kesehatan rambut. Penelitian ini mengevaluasi kinerja algoritma Naïve Bayes dalam klasifikasi tingkat rambut rontok menggunakan teknik data mining. Dataset yang digunakan terdiri dari 400 data individu dengan 10 atribut, termasuk atribut kelas yang menentukan tingkat rambut rontok (few, medium, many, a lot). Pengujian dilakukan menggunakan beberapa rasio data latih dan uji. Hasilnya menunjukkan akurasi tertinggi 76,67% pada rasio 70:30 dan terendah 55,00% pada rasio 90:10. Meskipun Naïve Bayes tidak secara langsung memberikan informasi atribut yang paling berpengaruh, analisis kepentingan atribut dapat diperoleh melalui pendekatan mutual information, uji chi-square, atau distribusi probabilitas atribut per kelas. Hasil penelitian ini memberikan gambaran mengenai potensi algoritma Naïve Bayes dalam mendukung sistem deteksi dini dan perawatan masalah rambut rontok secara lebih efektif. Kata Kunci: Rambut Rontok, Data Mining, Klasifikasi, Naïve Bayes vii ABSTRACT CLASSIFICATION OF HAIR LOSS USING THE NAÏVE BAYES METHOD By Arif Hidayat arifhidayat124@gmail.com Hair is an essential part of the human body, playing a crucial role in both appearance and health, and is therefore often a focus of personal care. Hair loss is one of the most common problems experienced by society, caused by various factors such as genetics and lifestyle. Based on a survey conducted by Lembaga Jajak Pendapat (Jakpat) in mid-2023 involving 3,041 respondents, the most prevalent hair issue was hair loss (64.7%), followed by dandruff (44.3%), dry and dull hair (30.8%), oily or greasy hair (26.1%), and split ends (18%). Most hair loss cases occurred in the 20–25 age group (37.7%). This problem highlights the need for early detection and proper treatment to maintain hair health. This study evaluates the performance of the Naïve Bayes algorithm in classifying hair loss levels using data mining techniques. The dataset consists of 400 individuals with 10 attributes, including a class attribute indicating hair loss levels (few, medium, many, a lot). Testing was conducted using multiple training-to-testing ratios. The results showed the highest accuracy of 76.67% at a 70:30 ratio and the lowest accuracy of 55.00% at a 90:10 ratio. Although Naïve Bayes does not directly provide feature importance like Decision Trees or Random Forests, attribute relevance can be analyzed through mutual information, chi-square tests, or probability distribution analysis per class. These findings illustrate the potential of the Naïve Bayes algorithm to support early detection systems and more effective hair loss treatment strategies. Key Words: Hair Loss, Data Mining, Classification, Naïve Bayes
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | Ilmu Komputer eTheses |
| Divisions: | Pasca Sarjana > Magister Teknik Informatika |
| Depositing User: | Arif Hidayat |
| Date Deposited: | 05 Nov 2025 03:13 |
| Last Modified: | 05 Nov 2025 03:13 |
| URI: | http://repo.darmajaya.ac.id/id/eprint/23033 |
Actions (login required)
 |
View Item |