Deep Learning dengan ResNet50 untuk Sistem Rekomendasi Fashion Berbasis Citra

Authors

  • Dewi Rahmawati Universitas Nusa Mandiri
  • Kanaya Salsabila Setiawan Universitas Nusa Mandiri
  • Muhammad Fahreza Reynaldy Universitas Nusa Mandiri
  • Rangga Ramadhan Universitas Nusa Mandiri

DOI:

https://doi.org/10.52958/iftk.v21i2.11967

Keywords:

Computer_Vision, Deep_Learning, Fashion, ResNet50, Sistem_Rekomendasi

Abstract

Perkembangan industri fashion yang pesat menuntut sistem rekomendasi yang tidak hanya akurat, tetapi juga mampu memahami preferensi visual pengguna. Sistem rekomendasi berbasis teks seringkali menghadapi keterbatasan dalam menangkap konteks visual yang kompleks, sehingga pendekatan berbasis citra menjadi solusi yang lebih relevan. Penelitian ini bertujuan untuk membangun sistem Smart Recommendation Search Engine berbasis visual dengan memanfaatkan model Convolutional Neural Network (CNN) ResNet50 sebagai feature extractor dan dataset DeepFashion. ResNet50 digunakan untuk mengekstraksi vektor fitur dari gambar produk fashion, yang kemudian dimanfaatkan dalam pencarian gambar serupa menggunakan algoritma K-Nearest Neighbors (KNN). Proses penelitian mencakup pra-pemrosesan data, ekstraksi fitur visual, pencarian kemiripan berbasis metrik kemiripan kosinus (cosine similarity), serta evaluasi sistem menggunakan metrik precision dan recall pada top-K results (hasil teratas). Hasil pengujian menunjukkan bahwa metrik cosine similarity memberikan performa terbaik dalam menemukan gambar dengan kemiripan visual tinggi, dengan nilai precision pada satu hasil teratas (precision at top-1) sebesar 0,230. Sistem yang dikembangkan berhasil mengidentifikasi produk fashion serupa secara visual dan mendukung pengalaman belanja yang lebih personal. Temuan ini menegaskan potensi pendekatan berbasis visual dalam meningkatkan akurasi sistem rekomendasi serta mendukung gaya hidup berkelanjutan.

References

A. Etuk, J. A. Anyadighibe, E. E. James, and M. U. Ukpe, “Sociological factors and consumer buying behaviour towards fashion clothing,” International Journal of Applied Research in Social Sciences, vol. 4, no. 2, pp. 21–34, 2022.

A. Balabanyts, I. Kyianytsia, T. Bielialov, S. Mosiiuk, O. Nedzvetska, and O. Verbytska, “The impact of tourism and digital media on the dynamics of the fashion industry in the global creative economy,” Multidisciplinary Reviews, vol. 7, 2024.

S. Chakraborty, M. S. Hoque, N. Rahman Jeem, M. C. Biswas, D. Bardhan, and E. Lobaton, “Fashion recommendation systems, models and methods: A review,” in Informatics, MDPI, 2021, p. 49.

O. K. Majeed, Z. ul Abideen, U. Arshad, R. H. Ali, A. Habib, and R. Mustafa, “Adaptivecloset: Reinforcement learning in personalized clothing recommendations,” in 2023 18th International Conference on Emerging Technologies (ICET), IEEE, 2023, pp. 305–309.

Q. Zhang, J. Lu, and Y. Jin, “Artificial intelligence in recommender systems,” Complex & intelligent systems, vol. 7, no. 1, pp. 439–457, 2021.

M. Mohiuddin Babu, S. Akter, M. Rahman, M. M. Billah, and D. Hack-Polay, “The role of artificial intelligence in shaping the future of Agile fashion industry,” Production Planning & Control, vol. 35, no. 15, pp. 2084–2098, 2024.

S. Silvester and S. Kurian, “Recommendation systems: enhancing personalization and customer experience,” in 2023 3rd International Conference on Smart Generation Computing, Communication and Networking (SMART GENCON), IEEE, 2023, pp. 1–6.

D.-N. Nguyen, V.-H. Nguyen, T. Trinh, T. Ho, and H.-S. Le, “A personalized product recommendation model in e-commerce based on retrieval strategy,” Journal of Open Innovation: Technology, Market, and Complexity, vol. 10, no. 2, p. 100303, 2024, doi: https://doi.org/10.1016/j.joitmc.2024.100303.

B. Yang, “Clothing design style recommendation using decision tree algorithm combined with deep learning,” Comput Intell Neurosci, vol. 2022, no. 1, p. 5745457, 2022.

C. Gao, X. Wang, X. He, and Y. Li, “Graph neural networks for recommender system,” in Proceedings of the fifteenth ACM international conference on web search and data mining, 2022, pp. 1623–1625.

M. Princessilia, J. Wijaya, W. Fernando, Y. Indrayudha, C. Martin, and R. Y. Ningsih, “Perancangan Aplikasi Berbasis AI untuk Rekomendasi Pakaian Harian sebagai Solusi Pembelian Impulsif,” Jurnal Sosial Teknologi, vol. 5, no. 1, pp. 112–122, 2025.

Z. Liu, P. Luo, S. Qiu, X. Wang, and X. Tang, “Deepfashion: Powering robust clothes recognition and retrieval with rich annotations,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 1096–1104. Accessed: Jul. 28, 2025. [Online]. Available: https://mmlab.ie.cuhk.edu.hk/projects/DeepFashion.html

M. K. Syiam, A. T. Wibowo, and E. B. Setiawan, “Fashion Recommendation System using Collaborative Filtering,” Building of Informatics Technology and Science (BITS), vol. 5, no. 2, 2023.

C.-C. Chang, C.-H. Wei, Y.-H. Wang, C.-H. T. Yang, and S. Hsiao, “Recommender System for Apparel Products Based on Image Recognition Using Convolutional Neural Networks,” Engineering Proceedings, vol. 89, no. 1, p. 38, 2025.

A. Çınar, M. Yıldırım, and Y. Eroğlu, “Classification of pneumonia cell images using improved ResNet50 model,” Traitement du Signal, vol. 38, no. 1, pp. 165–173, 2021.

B. E. Jin and D. C. Shin, “The power of 4th industrial revolution in the fashion industry: what, why, and how has the industry changed?,” Fashion and Textiles, vol. 8, no. 1, p. 31, 2021.

Z. Lu, Y. Hu, Y. Chen, and B. Zeng, “Personalized outfit recommendation with learnable anchors,” in Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 2021, pp. 12722–12731.

Published

2025-08-19

Issue

Section

INFORMATIK