<![CDATA[In recent years, the petshop industry has seen a significant increase. This is due to people's growing awareness of pet welfare and the need for specialized products for them. Product sales in petshops are not only influenced by customer preferences, but also by the diversity of pet breeds, which makes data collection and product stock management more complex. Therefore, predicting the products that are most in demand by customers is important. Data mining, as a part of computer science that focuses on extracting information from data, offers an effective way to analyze patterns and trends of product sales in petshops. In this study, Naive Bayes algorithm is used to predict the best-selling products in petshop. RapidMiner software was used to process the data in this study. Data processing with RapidMiner resulted in a prediction accuracy of 90.41%. Class precision for the prediction of hot-selling products is 88.24%, while for non-selling products is 90.00%. Class recall for the prediction of hot-selling products reaches 90.91%, while for products that are not in demand reaches 92.31%.]]>
Copyrights © 2024
