Implementation of Data Mining for Classification of Best-selling Electronic Products using the Naïve Bayes Algorithm at PT. Midea Planet Indonesia

• RIFKI PUTRA HARAHAP
• 2026

---

The development of information technology encourages companies to utilize data as a basis for strategic decision-making. Large amounts of stored sales data are often not optimally processed to produce valuable information. This study aims to apply the Naïve Bayes algorithm to classify the sales status of Microwave Oven (MCA) products at PT. Midea Planet Indonesia is categorized into best-selling and non-best-selling categories. The method used is Knowledge Discovery in Databases (KDD), which includes the stages of selection, preprocessing (cleaning, integration, reduction), transformation, data mining, and evaluation. The research dataset consists of 1,156 sales data from January to December 2024, which are then processed into 70 integrated datasets. The classification process is carried out manually in Microsoft Excel and retested in RapidMiner to ensure model consistency. The evaluation results using a confusion matrix show that the Naïve Bayes algorithm achieves 90.00% accuracy, 86.7% precision, and 97.50% recall. These results indicate that the model performs well at identifying product sales status. Thus, this method can support management decision-making in inventory planning and sales performance evaluation.

• Data Mining
• Classification
• Naive Bayes
• Sales
• RapidMiner

https://repository.bsi.ac.id/repo/82867/Implementation-of-Data-Mining-for-Classification-of-Best-selling-Electronic-Products-using-the-Naïve-Bayes-Algorithm-at-PT.-Midea-Planet-Indonesia
Anda dapat membaca file tugas akhir ini di repository.