Pengembangan Aplikasi Business Intelligence Berbasis Artificial Intelligence untuk Prediksi Keberhasilan Penjualan Produk dan Keberlanjutan UMKM
Year: 2025
Type: article
Abstract: Micro, Small, and Medium Enterprises (MSMEs) in Pesawaran Regency face significant challenges, particularly the reliance on traditional sales methods, limited innovation, and minimal use of information technology. Most entrepreneurs still determine sales strategies based on personal experience, often resulting in inaccurate sales forecasts and stock imbalances that disrupt financial stability and business continuity. To address this issue, an Artificial Intelligence based Business Intelligence application was developed using the Agile Software Development methodology. The system employs the Random Forest algorithm to predict product sales success. Evaluation results demonstrate that the model achieved an accuracy of 86%, with a precision of 0.88, recall of 0.93, and F1-score of 0.90, indicating strong predictive performance. To further improve system reliability, it is recommended to increase training data particularly in the “Failed” class and optimize Random Forest parameters, such as the number of decision trees, tree depth, and feature selection, to achieve better predictive outcomes .
Institution: Institut Teknologi dan Bisnis Diniyyah Lampung
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W7127056988