Collaborative Filtering Untuk Rekomendasi Produk Skincare Menggunakan Deep Learning

Qalbyassalam, Chaira (2022) Collaborative Filtering Untuk Rekomendasi Produk Skincare Menggunakan Deep Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 07211840000026-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (4MB)

Abstract

Skincare di Indonesia telah mengalami peningkatan selama beberapa tahun terakhir. Seiring berkembangnya waktu semakin banyak jenis produk skincare yang dijual di pasaran. Perkembangan ini diiringi pula dengan perkembangan teknologi yang memudahkan pengguna untuk membeli produk skincare secara online. Saat membeli produk skincare secara online pengguna skincare dapat mencari informasi produk dari ulasan pengguna lain yang telah menggunakan produk tersebut. Terdapat banyak ulasan yang ditulis banyak pengguna lain untuk setiap produk skincare, yang dapat digunakan sebagai pertimbangan pengguna sebelum membeli produk skincare. Namun, rating ulasan yang memiliki skala 1 sampai 5 dianggap belum cukup merepresentasikan kualitas sebuah produk, pengguna pun perlu membaca teks ulasan yang ditulis pengguna lain untuk mendapat informasi yang lebih spesifik mengenai kualitas produk tersebut. Banyaknya teks ulasan ini dapat membuat pengguna kewalahan untuk mengidentifikasi produk mana yang tepat ketika mereka berbelanja produk skincare serta menghabiskan banyak waktu untuk mengidentifikasi dan membaca setiap ulasan. Dalam hal ini sistem rekomendasi berperan untuk meningkatkan efisiensi pengguna dalam pengambilan keputusan sebelum membeli produk skincare. Oleh karena itu pada Tugas Akhir ini dikembangkan sebuah sistem rekomendasi produk skincare dengan menggunakan deep learning. Metode yang digunakan adalah Neural Collaborative Filtering (NCF) yaitu metode Collaborative Filtering dengan pendekatan neural network, metode ini akan dibandingkan dengan metode Collaborative Filtering biasa yaitu Matrix Factorization (MF). Hasil dari Tugas Akhir ini didapatkan performa model NCF cukup baik untuk mengembangkan sistem rekomendasi dengan RMSE sebesar 0.4931. Berdasarkan hasil, NCF ini juga memiliki performa 52.9% lebih baik dibanding MF.
==============================================================================================================================
Skincare in Indonesia has improved over the last few years. Over time, more and more types of skincare products are sold on the market. This development is also accompanied by technological developments that make it easier for users to buy skin care products online. When buying skin care products online, skin care users can find product information from reviews of other users who have used the product. There are many reviews written by many other users for each skin care product, which users can consider before buying a skin care product. However, rating reviews that have a scale of 1 to 5 are considered not sufficient to represent product quality, users need to read review texts written by other users to get more specific information about the quality of the product. This large amount of review text can make it difficult for users to identify the right product when they are shopping for skin care products and take a long time to identify and read each review. In this case, the recommendation system plays a role in increasing user efficiency in making decisions before buying skin care products. Therefore, in this research, a skin care product recommendation system was developed using deep learning. The method used is Neural Collaborative Filtering (NCF), namely the Collaborative Filtering method with a neural network approach, this method will be compared with the usual Collaborative Filtering method, namely Matrix Factorization (MF). The results of this final project show that the performance of the NCF model is good enough to develop a recommendation system with an RMSE of 0.4931. Based on the results, this NCF also has 52.9% better performance than MF.

Item Type:	Thesis (Other)
Additional Information:	RSKom 005.276 2 Qal c-1 2022
Uncontrolled Keywords:	Deep Learning. Produk Skincare. Sistem Rekomendasi. Deep Learning. Recommender System. Skincare Products.
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science. EDP
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	17 Jun 2026 01:59
Last Modified:	17 Jun 2026 01:59
URI:	http://repository.its.ac.id/id/eprint/133834

Actions (login required)

View Item