This study proposes a new hybrid deep learning algorithm for intrusion detection in IoT networks. The algorithm combines one-dimensional Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to detect DDoS attacks in a big data environment. The study uses the CICIoT2023 and TON_IoT datasets for training and testing. The datasets are preprocessed using feature selection based on correlation analysis to reduce the number of features and improve model performance. The hybrid algorithm is compared with ten traditional machine learning and deep learning algorithms, including Random Forest, Decision Tree, Gradient Boost, AdaBoost, Naive Bayes, Logistic Regression, K-Nearest Neighbor, CNN, MLP, and LSTM. The model's performance is evaluated using accuracy, precision, recall, and F1-score. The results show that the hybrid algorithm achieves a high accuracy rate of 99.995% for binary classification and 99.96% for multiclassification in the CICIoT2023 dataset. In the TON_IoT dataset, the binary classification success rate is 98.75%. The study also discusses the importance of intrusion detection in IoT networks, the challenges posed by DDoS attacks, and the effectiveness of the proposed hybrid algorithm in detecting and classifying attacks. The results demonstrate that the hybrid algorithm outperforms traditional methods in terms of accuracy and efficiency. The study contributes to the field of intrusion detection by developing a new hybrid deep learning algorithm that can effectively detect and classify DDoS attacks in IoT networks.This study proposes a new hybrid deep learning algorithm for intrusion detection in IoT networks. The algorithm combines one-dimensional Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) to detect DDoS attacks in a big data environment. The study uses the CICIoT2023 and TON_IoT datasets for training and testing. The datasets are preprocessed using feature selection based on correlation analysis to reduce the number of features and improve model performance. The hybrid algorithm is compared with ten traditional machine learning and deep learning algorithms, including Random Forest, Decision Tree, Gradient Boost, AdaBoost, Naive Bayes, Logistic Regression, K-Nearest Neighbor, CNN, MLP, and LSTM. The model's performance is evaluated using accuracy, precision, recall, and F1-score. The results show that the hybrid algorithm achieves a high accuracy rate of 99.995% for binary classification and 99.96% for multiclassification in the CICIoT2023 dataset. In the TON_IoT dataset, the binary classification success rate is 98.75%. The study also discusses the importance of intrusion detection in IoT networks, the challenges posed by DDoS attacks, and the effectiveness of the proposed hybrid algorithm in detecting and classifying attacks. The results demonstrate that the hybrid algorithm outperforms traditional methods in terms of accuracy and efficiency. The study contributes to the field of intrusion detection by developing a new hybrid deep learning algorithm that can effectively detect and classify DDoS attacks in IoT networks.