Can Kaymakci
Papers
2
Total Citations
8
H-Index
2
About
Can Kaymakci is a researcher at the forefront of industrial AI, specializing in anomaly detection, federated learning, and energy-efficient machine learning for manufacturing. Their work addresses critical challenges in applying AI to real-world industrial time series data, where data scarcity and privacy constraints often hinder deployment. Kaymakci’s most-cited paper, “A data-efficient active learning architecture for anomaly detection in industrial time series data” (2025, 5 citations), introduces a novel approach that reduces the need for labeled data—a major bottleneck in predictive maintenance—by intelligently selecting the most informative samples for human review. This work directly supports maintenance cost reduction and machine fault prevention. Their earlier contribution, “Federated Machine Learning Architecture for Energy-Efficient Industrial Applications” (2021, 3 citations), tackles the dual challenges of data privacy and energy consumption by enabling collaborative model training across decentralized factory sensors without sharing raw data. This architecture is pivotal for sustainable smart manufacturing, allowing companies to extract value from power consumption data while minimizing computational overhead. Kaymakci’s research bridges the gap between theoretical AI and practical industrial constraints, offering scalable, privacy-preserving solutions that are both data-efficient and energy-conscious—key achievements for the next generation of intelligent factories.
Research Focus
Key Achievements
Top Papers
- 1
- 2