Minority-Class Internet of Things Anomaly Detection via Deep Learning Models Using CESNET-TimeSeries24 Dataset
5th IEEE International Maghreb Meeting of the Conference on Sciences and Techniques of Automatic Control and Computer Engineering, MI-STA 2026, Sebha, Libya, 5 - 07 Nisan 2026, ss.385-392, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/mi-sta68962.2026.11511194
- Basıldığı Şehir: Sebha
- Basıldığı Ülke: Libya
- Sayfa Sayıları: ss.385-392
- Anahtar Kelimeler: Anomaly Detection, CESNET-TimeSeries24, Internet of Things, Intrusion Detection Systems, Long Short-Term Memory networks
- Ankara Yıldırım Beyazıt Üniversitesi Adresli: Evet
Özet
Accurate forecasting and anomaly detection in IoT networks are critical for maintaining secure and reliable operations. This study evaluates three deep learning architectures—Long Short-Term Memory (LSTM), Transformer, and Temporal Convolutional Network (TCN)—on the CESNET-TimeSeries24 dataset to assess their ability to capture complex temporal dependencies in network traffic. Comprehensive performance comparisons reveal that the LSTM model consistently achieves superior predictive accuracy, with a Mean Absolute Error (MAE) of 6.89×107, Mean Squared Error (MSE) of 4.44 × 1016, and Root Mean Squared Error (RMSE) of 2.11 × 108. In classification metrics, LSTM attains 82.31% accuracy, 81.15% precision, 92.92% recall, and an F1-score of 86.63%, outperforming both Transformer and TCN models. The Transformer model achieves the highest recall (97.06%) and the lowest RMSE (2.09 × 108), indicating robustness to large-magnitude errors, while TCN demonstrates a competitive F1-score (86.29%) with higher MAE. Visualizations of actual versus predicted values and training–validation loss curves confirm that LSTM provides the most stable and accurate predictions. These results highlight the effectiveness of LSTM for reliable IoT traffic forecasting and anomaly detection, balancing low regression error with high classification performance.