Forecasting Fashion Product Trends Using the SARIMA Method on Google Trends Data
Abstract
The development of the digital era is very rapid at this time there are so many technologies that are present in human life, one of which is e-commerce. Today's society is very dependent on what is called e-commerce, there are lots of transactions made every year from 2017 there are around 8,657 USD until 2021 which continues to increase to 53,808 USD. This is more or less influenced by advertisements scattered in cyberspace. In 2020, the 1st rank of the highest number of transactions is held by clothing products, so sellers from e-commerce need to be assisted with a recommendation system that displays the best time to display their advertisements. The process uses the time-series forecasting method using the SARIMA algorithm. This algorithm takes the trend of the season from the previous data, so it is very suitable for the cases mentioned earlier. The data retrieval processes up to deployment to the web system uses the CRISP-DM research method. The results of this study have a very good evaluation value, namely with an R2 of 0.36993 so that the model is good to use and in its implementation is a web system that takes the category and how long it will be predicted.
References
A. Aggarwal, M. Alshehri, M. Kumar, O. Alfarraj, P. Sharma, and K. R. Pardasani, “Landslide data analysis using various time-series forecasting models,” Computers and Electrical Engineering, vol. 88, Dec. 2020, doi: 10.1016/j.compeleceng.2020.106858.
A. Kumar Dubey, A. Kumar, V. García-Díaz, A. Kumar Sharma, and K. Kanhaiya, “Study and analysis of SARIMA and LSTM in forecasting time series data,” Sustainable Energy Technologies and Assessments, vol. 47, Oct. 2021, doi: 10.1016/j.seta.2021.101474.
A. Lidwina, “Produk Fesyen Jadi Primadona di E-Commerce | Databoks,” 2021. https://databoks.katadata.co.id/datapublish/2021/06/09/produk-fesyen-jadi-primadona-di-e-commerce
A. L. Kusumatrisna et al., Statistik E-Commerce 2020. 2020.
C. Siettou, “Societal interest in puppies and the Covid-19 pandemic: A google trends analysis,” Preventive Veterinary Medicine, vol. 196, Nov. 2021, doi: 10.1016/j.prevetmed.2021.105496.
F. Martinez-Plumed et al., “CRISP-DM Twenty Years Later: From Data Mining Processes to Data Science Trajectories,” IEEE Transactions on Knowledge and Data Engineering, vol. 33, no. 8, pp. 3048–3061, Aug. 2021, doi: 10.1109/TKDE.2019.2962680.
G. Reikard, “Forecasting paleoclimatic data with time series models,” Results in Geophysical Sciences, vol. 6, p. 100015, Jun. 2021, doi: 10.1016/j.ringps.2021.100015.
H. Budiman, “PERAN TEKNOLOGI INFORMASI DAN KOMUNIKASI DALAM PENDIDIKAN,” Al-Tadzkiyyah: Jurnal Pendidikan Islam, vol. 8, 2017.
H. Yakubu and C. K. Kwong, “Forecasting the importance of product attributes using online customer reviews and Google Trends,” Technological Forecasting and Social Change, vol. 171, Oct. 2021, doi: 10.1016/j.techfore.2021.120983.
K. He, L. Ji, C. W. D. Wu, and K. F. G. Tso, “Using SARIMA–CNN–LSTM approach to forecast daily tourism demand,” Journal of Hospitality and Tourism Management, vol. 49, pp. 25–33, Dec. 2021, doi: 10.1016/j.jhtm.2021.08.022.
K. Tanaka, “A sales forecasting model for new-released and nonlinear sales trend products,” in Expert Systems with Applications, Nov. 2010, vol. 37, no. 11, pp. 7387–7393. doi: 10.1016/j.eswa.2010.04.032.
Statista, “eCommerce - Indonesia | Statista Market Forecast,” 2021. https://www.statista.com/outlook/dmo/ecommerce/indonesia#revenue (accessed Jan. 15, 2022).
