Abstract: C2C (consumer-to-consumer) is a business model where two individuals transact or conduct business with each other using a platform. A consumer act as a seller put their product in a platform later will be displayed to another consumer act as a buyer. This condition encourages platform to maintain high quality product information especially image that is provided by the seller. Product images need to be relevant to the product itself. It can be controlled automatically using image classification. In this paper, we carried out a research to find out the best deep learning model in image classification for e-commerce products. A dataset of 12,500 product images is collected from various web sources to be used in training and testing process. Five models are selected and fine-tuned using a uniform hyperparameter set-up. Those hyperparameters are found by using a manual process by trying a lot of hyperparameters. The testing result from every model is presented and evaluated. The result shows that NASNetLarge yield the best performance among all evaluated models with 84% testing accuracy.

Herdian , Gede Putra Kusuma and Suharjito. “Classification of C2C e-Commerce Product Images using Deep Learning Algorithm”. International Journal of Advanced Computer Science and Applications (IJACSA) 10.9 (2019). http://dx.doi.org/10.14569/IJACSA.2019.0100925

@article{2019,
title = {Classification of C2C e-Commerce Product Images using Deep Learning Algorithm},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2019.0100925},
url = {http://dx.doi.org/10.14569/IJACSA.2019.0100925},
year = {2019},
publisher = {The Science and Information Organization},
volume = {10},
number = {9},
author = {Herdian and Gede Putra Kusuma and Suharjito}
}