<![CDATA[The retail sellers who use digital tools in business adaptation are can survive better during the pandemic. PT. Ayo Techno Idea, as a company that provides digital buying and selling transaction tools, has an application called Ngorder.id. Where application users (sellers) have difficulty dealing with uncertain stock needs manually. This study aims to explain the results of the implementation of the prediction method and dashboard visualization according to the needs of Ngorder.id application users. The research analysis procedure uses the Knowledge Discovery in Database process which consists of the data collection and evaluation stages. The results of the study have found that four data attributes are needed in the implementation of data mining prediction methods. Then, the results of the implementation of the prediction of product sales transactions using the Single Exponential Smoothing method got decent performance with details of one of the nine products having a MAPE value of 14.89%, seven products having a MAPE value range of 20% to 50%, and one product having a MAPE value of 68, 21%. While the implementation of the Double Exponential Smoothing method has a low-performance tendency. Where obtained seven products have a MAPE value of more than 50% and the remaining two have a MAPE value of 42.79% and 33.68%. In the study, the results of the visual representation of the data were divided into three pages with different focuses and got a usability score in the good category. So it is feasible to use it as a feature reference and to help make decisions on sales transactions for application products.]]>
Copyrights © 2022
