• B.Sharmila
• T.Udhayakumar

The extensive implementation of Internet of Things (IoT) applications has aided in the growth of smart cities. These cities employ smart applications to optimize operational efficiency, thereby improving service quality and public health. To reduce the risks associated with IoT cybersecurity in smart cities, I present in this research an attack and anomaly detection method based on machine learning techniques. Notably, I chose Convolutional Recurrent Neural Networks (CRNNs) from the various available machine learning (ML) techniques. Because CRNNs combine the best features of Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs), CNNs allow us to study sequential data and extract spatial characteristics, making them ideal for visual data analysis. RNNs, on the other hand, capture temporal dependencies and analyze sequential data. By Combining both CNNs and RNNs I can achieve comprehensive cybersecurity. Results from experiments using the most current attack dataset show that the suggested method can successfully detect cyberattacks and outperform findings from previous research.

Cybersecurity, ML, CNNs, threat detection, LSTM,1D Convolutional neural networks, K-Fold Cross validation, confusion Matrix, anomaly detection.

"Crnns for an effective cyber threat detection in smart cities", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.o284-o294, April-2024, Available :http://www.jetir.org/papers/JETIR2404F42.pdf

"Crnns for an effective cyber threat detection in smart cities", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppo284-o294, April-2024, Available at : http://www.jetir.org/papers/JETIR2404F42.pdf