Larasati, Aisya (2026) Sistem Rekomendasi Produk Skincare Pada Toko Berlian Indah Menggunakan Metode Based Filtering Berbasis Website. S1 Teknik Informatika thesis, STMIK Widya Cipta Dharma.
![[img]](https://repository.wicida.ac.id/style/images/fileicons/text.png) |
Text
2243035-S1-Teknik Informatika.pdf Restricted to Repository staff only Download (3MB) | Request a copy |
|
Text
2243035-S1-Jurnal.pdf Download (428kB) |
Abstract
Perkembangan industri skincare yang semakin pesat menyebabkan banyaknya variasi produk yang tersedia di pasaran, sehingga konsumen sering mengalami kesulitan dalam menentukan produk yang sesuai dengan jenis kulit dan kebutuhan perawatan kulit. Permasalahan yang sama juga terjadi pada Toko Berlian Indah, di mana proses pemberian rekomendasi produk masih dilakukan secara manual berdasarkan pengetahuan penjual sehingga membutuhkan waktu dan belum memberikan rekomendasi secara optimal. Oleh karena itu, diperlukan suatu sistem rekomendasi yang mampu membantu pelanggan memperoleh produk skincare yang sesuai dengan karakteristik dan kebutuhan kulit mereka. Penelitian ini bertujuan untuk membangun sistem rekomendasi produk skincare berbasis website menggunakan metode Content Based Filtering. Metode tersebut diterapkan melalui fitur Pertanyaan Rekomendasi, di mana pengguna memilih jenis kulit dan kategori produk yang diinginkan sebagai preferensi awal. Selanjutnya, sistem melakukan pembobotan atribut produk menggunakan metode Term Frequency-Inverse Document Frequency (TF-IDF) dan menghitung tingkat kemiripan menggunakan Cosine Similarity berdasarkan atribut jenis kulit, kandungan, manfaat, dan kategori produk sehingga menghasilkan rekomendasi produk dengan tingkat kemiripan tertinggi. Sistem dikembangkan menggunakan framework Laravel, bahasa pemrograman PHP, dan database MySQL dengan data sebanyak 300 produk skincare yang diperoleh dari Toko Berlian Indah. Hasil penelitian menunjukkan bahwa sistem rekomendasi yang dikembangkan mampu memberikan rekomendasi produk skincare sesuai dengan preferensi pengguna berdasarkan metode Content Based Filtering. Berdasarkan hasil Black Box Testing, seluruh fungsi pada website berjalan sesuai dengan kebutuhan sistem. Hasil White Box Testing menunjukkan bahwa logika program pada proses pembentukan vektor TF-IDF dan perhitungan Cosine Similarity telah berjalan sesuai dengan alur yang dirancang. Selain itu, hasil Beta Testing yang melibatkan 30 responden menunjukkan bahwa website mudah digunakan, informasi produk mudah dipahami, serta fitur rekomendasi mampu membantu pelanggan dalam menentukan produk skincare yang sesuai dengan jenis kulit dan kebutuhan mereka. Dengan demikian, sistem rekomendasi produk skincare berbasis website yang dibangun dapat menjadi media pendukung bagi Toko Berlian Indah dalam meningkatkan kualitas pelayanan kepada pelanggan melalui penyediaan rekomendasi produk yang lebih efektif, efisien, dan sesuai dengan kebutuhan pengguna. ============================================================ The rapid growth of the skincare industry has resulted in a wide variety of skincare products being available on the market, making it difficult for consumers to select products that match their skin type and skincare needs. A similar problem occurs at Berlian Indah Store, where product recommendations are still provided manually based on the seller's knowledge, making the recommendation process time-consuming and less effective. Therefore, a recommendation system is needed to assist customers in selecting skincare products that best suit their skin characteristics and personal needs. This study aims to develop a website-based skincare product recommendation system using the Content Based Filtering method. The method is implemented through the Recommendation Questions feature, where users select their skin type and preferred product category as their initial preferences. The system then applies Term Frequency-Inverse Document Frequency (TF-IDF) for product attribute weighting and Cosine Similarity to calculate the similarity between user preferences and product attributes, including skin type, ingredients, benefits, and product category, in order to generate the most relevant product recommendations. The system was developed using the Laravel framework, PHP programming language, and MySQL database with a dataset consisting of 300 skincare products collected from Berlian Indah Store. The results indicate that the developed recommendation system is capable of providing skincare product recommendations that match user preferences using the Content Based Filtering method. Black Box Testing shows that all website functions operate according to the system requirements, while White Box Testing confirms that the program logic for TF-IDF vector generation and Cosine Similarity calculations performs as designed. Furthermore, Beta Testing involving 30 respondents demonstrates that the website is easy to use, product information is easy to understand, and the recommendation feature effectively assists customers in selecting skincare products according to their skin type and skincare needs. Therefore, the developed website-based skincare recommendation system can support Berlian Indah Store in improving customer service by providing more effective, efficient, and relevant skincare product recommendations.
| Item Type: | Thesis (S1 Teknik Informatika) |
|---|---|
| Additional Information: | Pembimbing 1 : Drs. Azahari., M.Kom Pembimbing 2 : Ulfah Nurfadhila., M.Pd |
| Uncontrolled Keywords: | Content Based Filtering, TF-IDF, Cosine Similarity, Sistem Rekomendasi, Website. |
| Subjects: | Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Teknik Informatika |
| Depositing User: | Ms Aisya Larasati |
| Date Deposited: | 06 Aug 2026 08:41 |
| Last Modified: | 06 Aug 2026 08:41 |
| URI: | http://repository.wicida.ac.id/id/eprint/6402 |
Actions (login required)
 |
View Item |
