Preferred Language
Articles
/
1ubFxZ0BmraWrQ4dNlu7
A Robust Base-layer Design for Hierarchical IoT Intrusion Detection Using Hybrid Deep Learning
...Show More Authors

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 Grey Wolf Optimization (GWO) with Borderline-SMOTE. Particle Swarm Optimization (PSO) was used to select the most important features that provide the greatest amount of information for training the first layer, which was built using deep learning techniques, and linking them in a hybrid manner that combines a Convolutional Neural Network–Long Short-Term Memory (CNN–LSTM) and the Attention mechanism. The proposed model was evaluated using two different types of datasets: the CICIOT2023 dataset, which is characterized by its large size and significant variation in the number of attacks, and the UNSW-NB15 dataset, which is characterized by its simplicity and less imbalance compared to the first dataset, to prepare and generalize the system across multiple environments. The proposed class showed binary classification results with an accuracy of 0.94, an Area Under the Curve (AUC) of 0.93, an optimized F1-score of 0.338, and a Matthews Correlation Coefficient (MCC) of 0.324 at the best threshold on the CICIoT2023 dataset. It also achieved an accuracy of approximately 0.96, an AUC of 0.985, and an MCC of over 0.82 on the UNSW-NB15 dataset. These results confirmed the construction of a strong and resilient layer, preparing the foundation for a robust hierarchical offside detection system.

Scopus
View Publication Preview PDF
Quick Preview PDF
Publication Date
Thu Jan 01 2015
Journal Name
Energy Sources, Part A: Recovery, Utilization, And Environmental Effects
Ultra Deep Hydrotreatment of Iraqi Vacuum Gas Oil Using a Modified Catalyst
...Show More Authors

A set of hydro treating experiments are carried out on vacuum gas oil in a trickle bed reactor to study the hydrodesulfurization and hydrodenitrogenation based on two model compounds, carbazole (non-basic nitrogen compound) and acridine (basic nitrogen compound), which are added at 0–200 ppm to the tested oil, and dibenzotiophene is used as a sulfur model compound at 3,000 ppm over commercial CoMo/ Al2O3 and prepared PtMo/Al2O3. The impregnation method is used to prepare (0.5% Pt) PtMo/Al2O3. The basic sites are found to be very small, and the two catalysts exhibit good metal support interaction. In the absence of nitrogen compounds over the tested catalysts in the trickle bed reactor at temperatures of 523 to 573 K, liquid hourly space v

... Show More
Publication Date
Tue Apr 30 2024
Journal Name
Iraqi Journal Of Science
Crescent Moon Visibility: A New Criterion using Deep learned Artificial Neural-Network
...Show More Authors

     Many authors investigated the problem of the early visibility of the new crescent moon after the conjunction and proposed many criteria addressing this issue in the literature. This article presented a proposed criterion for early crescent moon sighting based on a deep-learned pattern recognizer artificial neural network (ANN) performance. Moon sight datasets were collected from various sources and used to learn the ANN. The new criterion relied on the crescent width and the arc of vision from the edge of the crescent bright limb. The result of that criterion was a control value indicating the moon's visibility condition, which separated the datasets into four regions: invisible, telescope only, probably visible, and certai

... Show More
Preview PDF
Scopus (2)
Crossref (2)
Scopus Crossref
Publication Date
Wed Mar 16 2022
Journal Name
2022 Muthanna International Conference On Engineering Science And Technology (micest)
A hybrid feature selection technique using chi-square with genetic algorithm
...Show More Authors

View Publication
Scopus (2)
Crossref (2)
Scopus Crossref
Publication Date
Tue Nov 03 2020
Journal Name
Modern Sport
Effect of using flat hierarchical method on some aspects of muscular strength and straight serve skill for nascent tennis players
...Show More Authors

The researcher focused on the importance of the physical abilities of the tennis game, as this game is one of the games that are characterized by its specificity in performance as this game is characterized by continuous movement and dealing with different elements, so this game requires the development of muscle strength, which plays an important role in Performance skills in the game of tennis. There are several methods to develop strength, including flat hierarchical technique, which is one of the most common forms of training in the development of muscle strength.     As for the research problem, the researcher found a method that has an effect on the development of force. Therefore, the researcher tried to diversify a

... Show More
View Publication Preview PDF
Crossref (1)
Crossref
Publication Date
Tue Dec 12 2017
Journal Name
Al-khwarizmi Engineering Journal
Design a Vegetative Filter Strips Length Using VFSMOD_W Model for Reducing Sediments and Pesticides
...Show More Authors

 Abstract 

The vegetative filter strips (VFS) are a useful tool used for reducing the movement of sediment and pesticide in therivers. The filter strip’s soil can help in reducing the runoff volume by infiltration. However, the characteristics of VFS (i.e., length) are not recently identified depending on the estimation of VFS modeling performance. The aim of this research is to study these characteristics and determine acorrelation between filter strip length and percent reduction (trapping efficiency) for sediment, water, and pesticide. Two proposed pesticides(one has organic carbon sorption coefficient, Koc, of 147 L/kg which is more moveable than XXXX, and another one

... Show More
View Publication Preview PDF
Crossref
Publication Date
Wed Dec 18 2019
Journal Name
Baghdad Science Journal
Eye Detection using Helmholtz Principle
...Show More Authors

            Eye Detection is used in many applications like pattern recognition, biometric, surveillance system and many other systems. In this paper, a new method is presented to detect and extract the overall shape of one eye from image depending on two principles Helmholtz & Gestalt. According to the principle of perception by Helmholz, any observed geometric shape is perceptually "meaningful" if its repetition number is very small in image with random distribution. To achieve this goal, Gestalt Principle states that humans see things either through grouping its similar elements or recognize patterns. In general, according to Gestalt Principle, humans see things through genera

... Show More
View Publication Preview PDF
Scopus (8)
Crossref (3)
Scopus Clarivate Crossref
Publication Date
Tue Mar 31 2020
Journal Name
International Journal Of Heat And Technology
Enhancement of Natural Convection Heat Transfer of Hybrid Design Heat Sink
...Show More Authors

View Publication
Scopus (3)
Crossref (2)
Scopus Clarivate Crossref
Publication Date
Wed Oct 21 2015
Journal Name
Integrated Journal Of Engineering Research And Technology
A HYBRID CUCKOO SEARCH AND BACK-PROPAGATION ALGORITHMS WITH DYNAMIC LEARNING RATE TO SPEED UP THE CONVERGENCE (SUBPL) ALGORITHM
...Show More Authors

BP algorithm is the most widely used supervised training algorithms for multi-layered feedforward neural net works. However, BP takes long time to converge and quite sensitive to the initial weights of a network. In this paper, a modified cuckoo search algorithm is used to get the optimal set of initial weights that will be used by BP algorithm. And changing the value of BP learning rate to improve the error convergence. The performance of the proposed hybrid algorithm is compared with the stan dard BP using simple data sets. The simulation result show that the proposed algorithm has improved the BP training in terms of quick convergence of the solution depending on the slope of the error graph.

Publication Date
Fri Aug 12 2022
Journal Name
Future Internet
Improved DDoS Detection Utilizing Deep Neural Networks and Feedforward Neural Networks as Autoencoder
...Show More Authors

Software-defined networking (SDN) is an innovative network paradigm, offering substantial control of network operation through a network’s architecture. SDN is an ideal platform for implementing projects involving distributed applications, security solutions, and decentralized network administration in a multitenant data center environment due to its programmability. As its usage rapidly expands, network security threats are becoming more frequent, leading SDN security to be of significant concern. Machine-learning (ML) techniques for intrusion detection of DDoS attacks in SDN networks utilize standard datasets and fail to cover all classification aspects, resulting in under-coverage of attack diversity. This paper proposes a hybr

... Show More
View Publication Preview PDF
Scopus (44)
Crossref (36)
Scopus Clarivate Crossref
Publication Date
Tue Dec 05 2023
Journal Name
Baghdad Science Journal
AlexNet-Based Feature Extraction for Cassava Classification: A Machine Learning Approach
...Show More Authors

Cassava, a significant crop in Africa, Asia, and South America, is a staple food for millions. However, classifying cassava species using conventional color, texture, and shape features is inefficient, as cassava leaves exhibit similarities across different types, including toxic and non-toxic varieties. This research aims to overcome the limitations of traditional classification methods by employing deep learning techniques with pre-trained AlexNet as the feature extractor to accurately classify four types of cassava: Gajah, Manggu, Kapok, and Beracun. The dataset was collected from local farms in Lamongan Indonesia. To collect images with agricultural research experts, the dataset consists of 1,400 images, and each type of cassava has

... Show More
View Publication Preview PDF
Scopus (10)
Crossref (5)
Scopus Crossref