Effect of Epoch on Accuracy using Convolutional Neural Network to Classification of fashion and furniture
DOI:
https://doi.org/10.29408/jit.v5i1.4393Kata Kunci:
CNN, fashion, furniture, epochAbstrak
Fashion and furniture are a necessity for every individual. The choice of fashion can sometimes also be someone else's assessment of the character possessed by each individual. Likewise, the selection of furniture can also describe the identity of each individual. However, fashion and furniture have various models, for that we need information technology that can distinguish between model a and model b. Classification is a model that groups based on the same criteria. The use of classification must be supported by the right method and the use of epochs so that the resulting accuracy can be maximized. The Convolutional Neural Network (CNN) method is a very appropriate method for image classification because there are many architectures that can be used. The results of this study indicate that the best accuracy in furniture = 94.18% with the use of epoch = 500, and fashion = 99.15% with the use of epoch = 1500.Referensi
A. A. Riyanto, “Sejarah Dan Perkembangan Mode Busana,” 2005.
C. Candra, A. Santosa, and M. T. Rizqy, “Perancangan Mebel Multifungsi untuk ‘ Daily Treats ’ Surabaya,” J. Intra, vol. 5, no. 2, pp. 322–331, 2017.
M. S. Wibawa, “Pengaruh Fungsi Aktivasi , Optimisasi dan Jumlah Epoch Terhadap Performa Jaringan Saraf Tiruan,” J. Sist. dan Inform., vol. 11, no. 2, pp. 167–174, 2017, doi: 10.13140/RG.2.2.21139.94241.
A. Bengio, Yoshua, Goodfellow.ian, Courville, “Deep learning,” Nature, vol. 29, no. 7553, pp. 1–73, 2015.
D. Justus, J. Brennan, S. Bonner, and A. S. McGough, “Predicting the Computational Cost of Deep Learning Models,” Proc. - 2018 IEEE Int. Conf. Big Data, Big Data 2018, pp. 3873–3882, 2019, doi: 10.1109/BigData.2018.8622396.
Y. Kanada, “Optimizing neural-network learning rate by using a genetic algorithm with per-epoch mutations,” Proc. Int. Jt. Conf. Neural Networks, vol. 2016-Octob, no. March, pp. 1472–1479, 2016, doi: 10.1109/IJCNN.2016.7727372.
Y. You, Z. Zhang, C. J. Hsieh, J. Demmel, and K. Keutzer, “ImageNet training in minutes,” ACM Int. Conf. Proceeding Ser., no. September, 2018, doi: 10.1145/3225058.3225069.
N. Inoue, E. Simo-Serra, T. Yamasaki, and H. Ishikawa, “Multi-label Fashion Image Classification with Minimal Human Supervision,” Proc. - 2017 IEEE Int. Conf. Comput. Vis. Work. ICCVW 2017, vol. 2018-Janua, pp. 2261–2267, 2017, doi: 10.1109/ICCVW.2017.265.
W. Wang, Y. Xu, J. Shen, and S. C. Zhu, “Attentive Fashion Grammar Network for Fashion Landmark Detection and Clothing Category Classification,” Proc. IEEE Comput. Soc. Conf. Comput. Vis. Pattern Recognit., pp. 4271–4280, 2018, doi: 10.1109/CVPR.2018.00449.
A. Kimura, I. Takahashi, M. Tanaka, N. Yasuda, N. Ueda, and N. Yoshida, “Single-Epoch Supernova Classification with Deep Convolutional Neural Networks,” Proc. - IEEE 37th Int. Conf. Distrib. Comput. Syst. Work. ICDCSW 2017, pp. 354–359, 2017, doi: 10.1109/ICDCSW.2017.47.
Y. Seo and K. shik Shin, “Hierarchical convolutional neural networks for fashion image classification,” Expert Syst. Appl., vol. 116, pp. 328–339, 2019, doi: 10.1016/j.eswa.2018.09.022.
Sudianto Aris, “Penerapan Website Sebagai Sarana Promosi Wisata Budaya pada Kabupaten Lombok Timur,” Infotek J. Inform. dan Teknol., vol. 1Sudianto, no. 1, pp. 11–17, 2018.
A. Sudianto and H. Ahmadi, “Rancang Bangun Sistem Informasi Penjualan Sparepart Motor Pada Bengkel Vinensi Motor Berbasis Web Guna Meningkatkan Penjualan dan Promosi Produk Pendahuluan Vinensi sepeda motor motor,” Infotek J. Inform. dan Teknol., vol. 3, no. 2, pp. 32–39, 2020.
A. Sudianto and J. Sugiantara, “Website as Foundation Information Media under the auspices of Nahdlatul Wathan,” J. Phys. Conf. Ser., vol. 1539, no. 1, pp. 3–8, 2020, doi: 10.1088/1742-6596/1539/1/012024
S. M. Sudianto Aris, “Penerapan Sistem Informasi Geografis (GIS) dalam Pemetaan Kerajinan Kain Tenun dan Gerabah untuk Meningkatkan Potensi Kerajinan di Kabupaten Lombok Timur,” Infotek J. Inform. dan Teknol. J. Inform. dan Teknol., vol. 1, no. 2, pp. 64–71, 2018.
Unduhan
Diterbitkan
Cara Mengutip
Terbitan
Bagian
Lisensi
Semua tulisan pada jurnal ini menjadi tanggung jawab penuh penulis. Jurnal Infotek memberikan akses terbuka terhadap siapapun agar informasi dan temuan pada artikel tersebut bermanfaat bagi semua orang. Jurnal Infotek ini dapat diakses dan diunduh secara gratis, tanpa dipungut biaya sesuai dengan lisense creative commons yang digunakan.Jurnal Infotek is licensed under a Creative Commons Attribution 4.0 International License.
Statistik Pengunjung