Nowadays, internet security is a critical concern; the One of the most difficult study issues in network security is "intrusion detection". Fight against external threats. Intrusion detection is a novel method of securing computers and data networks that are already in use. To boost the efficacy of intrusion detection systems, machine learning and deep learning are widely deployed. While work on intrusion detection systems is already underway, based on data mining and machine learning is effective, it requires to detect intrusions by training static batch classifiers regardless considering the time-varying features of a regular data stream. Real-world problems, on the other hand, rarely fit into models that have such constraints. Furthermor

The rapid development of Internet of Things (IoT) devices and their increasing numbers have caused a tremendous increase in network traffic and a wider range of cyber-attacks. This growing trend has complicated the detection process for traditional intrusion detection systems and heightened the challenges faced by these devices, such as imbalanced and large training data. This study presents a cohesive methodology of a series of intelligent techniques to prepare clean and balanced data for training the first (core) layer of a robust hierarchical intrusion detection system. The methodology was built by cleaning and compressing the data using an Autoencoder and preparing a strong latent space for balancing using a hybrid method that combines

The Internet of Things (IoT) has significantly transformed modern systems through extensive connectivity but has also concurrently introduced considerable cybersecurity risks. Traditional rule-based methods are becoming increasingly insufficient in the face of evolving cyber threats.  This study proposes an enhanced methodology utilizing a hybrid machine-learning framework for IoT cyber-attack detection. The framework integrates a Grey Wolf Optimizer (GWO) for optimal feature selection, a customized synthetic minority oversampling technique (SMOTE) for data balancing, and a systematic approach to hyperparameter tuning of ensemble algorithms: Random Forest (RF), XGBoost, and CatBoost. Evaluations on the RT-IoT2022 dataset demonstrat

With the development of communication technologies for mobile devices and electronic communications, and went to the world of e-government, e-commerce and e-banking. It became necessary to control these activities from exposure to intrusion or misuse and to provide protection to them, so it's important to design powerful and efficient systems-do-this-purpose. It this paper it has been used several varieties of algorithm selection passive immune algorithm selection passive with real values, algorithm selection with passive detectors with a radius fixed, algorithm selection with passive detectors, variable- sized intrusion detection network type misuse where the algorithm generates a set of detectors to distinguish the self-samples. Practica

In recent years, there has been expanding development in the vehicular part and the number of vehicles moving on the roads in all the sections of the country. Arabic vehicle number plate identification based on image processing is a dynamic area of this work; this technique is used for security purposes such as tracking of stolen cars and access control to restricted areas. The License Plate Recognition System (LPRS) exploits a digital camera to capture vehicle plate numbers is used as input to the proposed recognition system. Basically, the proposed system consists of three phases, vehicle license plate localization, character segmentation, and character recognition, the License Plate (LP) detection is presented using canny edge detection

This research describes a new model inspired by Mobilenetv2 that was trained on a very diverse dataset. The goal is to enable fire detection in open areas to replace physical sensor-based fire detectors and reduce false alarms of fires, to achieve the lowest losses in open areas via deep learning. A diverse fire dataset was created that combines images and videos from several sources. In addition, another self-made data set was taken from the farms of the holy shrine of Al-Hussainiya in the city of Karbala. After that, the model was trained with the collected dataset. The test accuracy of the fire dataset that was trained with the new model reached 98.87%.

Recently, detecting dynamic patterns for growing communities in social networks have attracted significant attention. The objective of dynamic community detection is to analyze and identify clusters in complex networks that change over time. Different optimization algorithms, including both single-objective and multi-objective approaches, have been employed to address the challenge of dynamic community detection. Although current evolutionary algorithms for identifying community structure can traverse extensive areas of partition space, they often become stuck in local minima. In addition, limited studies have addressed this issue by integrating local search strategies with evolutionary algorithms for community identification. This paper in

Modern machine-learning applications require GPUs, and modern platforms can leverage numerous GPUs on one or more machines to increase performance. Contemporary deep-learning models are too huge for CPU or GPU training. Training these models with many GPUs without performance degradation is necessary to train them rapidly and maximize GPU consumption. Thus, training deep convolutional neural networks (DCNN) with multiple GPUs has become necessary for improving training. Therefore, we presented a parallel design and development of an efficient model for enhancing face mask CNN performance and improving resource efficiency. This DCNN model is a parallel training system over multiple GPUs, a multi-core CPU, and a multi-process GPU platform wit