<![CDATA[The Bombang Panas Farmers Group in Wue Village, Wolomeze District, Ngada Regency, produces various vegetable commodities. However, the sales process is still conducted through local markets, direct buyers, and WhatsApp, resulting in unstructured order records, limited marketing reach, unclear delivery addresses, and potential payment delays. This study aims to develop a web-based agricultural product sales information system to support the management of products, orders, and transactions, as well as to apply the Naïve Bayes algorithm to classify vegetable sales levels. The system was developed using the Waterfall model, which consists of requirements analysis, system design, implementation, testing, and maintenance. Data were collected through observation, interviews, and literature study. The classification dataset consists of five records with product, price, harvest quantity, and sales category attributes. The Naïve Bayes algorithm was applied by calculating the probability of each class based on the attributes of the classified data. The calculation for Kangkung, with a low price and large harvest quantity, produced a probability value of 0.20 for the High Sales class, while the Medium Sales and Low Sales classes each obtained a value of 0.00. Based on these results, Kangkung was classified as High Sales. Black Box Testing showed that the main system functions, including login, product management, ordering, payment, and classification, operated according to the testing scenarios. Functional system testing was distinguished from algorithm accuracy measurement because the available dataset did not provide an adequate independent test set.]]>
Copyrights © 2026
