Application of Machine Learning for Anomaly Detection in Printed Circuit Boards Imbalance Date Set

Abstract

The detection of anomalies in printed circuit boards (PCBs) is an important challenge in the electronics manufacturing industry. Traditional anomaly detection methods often struggle to handle imbalanced datasets, which are common in real-world PCB production. In recent years, machine learning (ML) algorithms have emerged as a promising solution to this problem. This study investigates the use of ML algorithms for anomaly detection in PCBs, with a particular focus on addressing the issue of imbalanced data. We propose a data-level technique to balance the dataset and improve the performance of the ML algorithm. Our results show that our approach outperforms traditional methods in terms of precision, recall, and F1 score. Overall, this study demonstrates the potential of ML in addressing the challenge of anomaly detection in PCBs and highlights the importance of considering imbalanced data in such applications.