Year: 2024
Type: article
Abstract: Naive Bayes algorithm used to classify the best-selling product categories in e-commerce. The data used comes from a public Kaggle dataset, comprising 250,000 transactions during the 2020–2023 period. The analysis process follows the CRISP-DM model, including stages such as business understanding, data preparation, modeling, and model evaluation using a confusion matrix. Evaluation results show that the model achieved an accuracy of 92.64%, precision of 91.51%, and recall of 96.68%. The analysis revealed that the best-selling product category is Clothing, followed by Books, Electronics, and Home. This study demonstrates that the Naive Bayes algorithm can be effectively implemented to support stock management and data-driven marketing strategies in e-commerce.
Source: Jurnal Sistem Informasi (JUSIN)
Institution: Universitas Katolik Musi Charitas
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W4412730745