Your search

Over recent decades, research in Artificial Intelligence (AI) has developed a broad range of approaches and methods that can be utilized or adapted to address complex optimization problems. As real-world problems get increasingly complicated, this requires an effective optimization method. Various meta-heuristic algorithms have been developed and applied in the optimization domain. This paper used and ameliorated a promising meta-heuristic approach named Crow Search Algorithm (CSA) to address numerical optimization problems. Although CSA can efficiently optimize many problems, it needs more searchability and early convergence. Its positioning updating process was improved by supporting two adaptive parameters: flight length (fl) and awareness probability (AP) to tackle these curbs. This is to manage the exploration and exploitation conducts of CSA in the search space. This process takes advantage of the randomization of crows in CSA and the adoption of well-known growth functions. These functions were recognized as exponential, power, and S-shaped functions to develop three different improved versions of CSA, referred to as Exponential CSA (ECSA), Power CSA (PCSA), and S-shaped CSA (SCSA). In each of these variants, two different functions were used to amend the values of fl and AP. A new dominant parameter was added to the positioning updating process of these algorithms to enhance exploration and exploitation behaviors further. The reliability of the proposed algorithms was evaluated on 67 benchmark functions, and their performance was quantified using relevant assessment criteria. The functionality of these algorithms was illustrated by tackling four engineering design problems. A comparative study was made to explore the efficacy of the proposed algorithms over the standard one and other methods. Overall results showed that ECSA, PCSA, and SCSA have convincing merits with superior performance compared to the others.

This paper introduces an enhanced version of the Capuchin Search Algorithm (CapSA) called ECapSA. CapSA draws inspiration from the collective intelligence of Capuchin monkeys and has shown success in solving real-world problems. However, it may encounter challenges handling complex optimization tasks, such as premature convergence or being trapped in local optima. ECapSA employs a local escaping mechanism operating the abandonment limit concept to exploit potential solutions and introduce diversification trends. Additionally, the ECapSA algorithm is improved by integrating the principles of the cooperative island model, resulting in the iECapSA. This modification enables better management of population diversity and a more optimal balance between exploration and exploitation. The efficiency of iECapSA is validated through a series of experiments, including the IEEE-CEC2014 benchmark functions and training the feedforward neural network (FNN) on seven biomedical datasets. The performance of iECapSA is compared to other metaheuristic techniques, namely differential evolution (DE), sine cosine algorithm (SCA), and whale optimization algorithm (WOA). The results of the comparative study demonstrate that iECapSA is a strong contender and surpasses other training algorithms in most datasets, particularly in terms of its ability to avoid local optima and its improved convergence speed.

In this study, we conducted experiments to model the temperature of two manufacturing processes using various metaheuristic search algorithms. The two processes adopted were the P05 horny steel tool and the AISI304 stainless steel castings machines. Our approach involves building a data-driven model, as traditional search methods for modeling manufac-turing problems often need help finding the global optimum when faced with a complex objective function and numerous decision variables. Bio-inspired metaheuristic search algorithms have shown promising performance in handling multi-model optimization functions, and efficiently exploring the search space to attain more global results. We applied several metaheuristic search algorithms to find the optimal tuning parameters of a temperature-based model. The results from the case studies demonstrate that Particle Swarm Optimization (PSO) provided the best performance in tuning model parameters, resulting in minimum modeling error.

Sleep apnea is a sleeping disorder affecting more than 20 % of all American adults, associated with intermittent air passageway obstruction during sleep. This results in intermittent hypoxia, sympathetic activation, and an interruption of sleep with various health consequences. The diagnosis of sleep apnea traditionally involves the performance of overnight polysomnography, where oxygen, heart rate, and breathing, among other physiologic variables, are continuously monitored during sleep at a sleep center. However, these sleep studies are expensive and impose access issues, given the number of patients who need to be diagnosed. There is hence utility in having an effective triage system to screen for OSA to utilize polysomnography better. In this study, we plan to explore using several machine learning algorithms to utilize pre-screening symptoms to diagnose obstructive sleep apnea (OSA). Per our experimental results, it was found that Decision Tree Classifier (DTC) and Random Forest (RF) provided the highest classification accuracies compared to other algorithms such as Logistic Regression (LR), Support Vector Machines (SVM), Gradient Boosting Classifier (GBC), Gaussian Naive Bayes (GNB), K Neighbors Classifier (KNC), and Artificial Neural Networks (ANN).

Identification of the optimal subset of features for Feature Selection (FS) problems is a demanding problem in machine learning and data mining. A trustworthy optimization approach is required to cope with the concerns involved in such a problem. Here, a Binary version of the Capuchin Search Algorithm (CSA), referred to as BCSA, was developed to select the optimal feature combination. Owing to the imbalance of parameters and random nature of BCSA, it may sometimes fall into the trap of an issue called local maxima. To beat this problem, the BCSA could be further improved with the resettlement of its individuals by adopting some methods of repopulating the individuals during foraging. Lévy flight was applied to augment the exploitation and exploration abilities of BCSA, a method referred to as LBCSA. A Chaotic strategy is used to reinforce search behavior for both exploration and exploitation potentials of BCSA, which is referred to as CBCSA. Finally, Lévy flight and chaotic sequence are integrated with BCSA, referred to as LCBCSA, to increase solution diversity and boost the openings of finding the global optimal solutions. The proposed methods were assessed on twenty-six datasets collected from the UCI repository. The results of these methods were compared with those of other FS methods. Overall results show that the proposed methods render more precise solutions in terms of accuracy rates and fitness scores than other methods.

Winning football matches is the major goal of all football clubs in the world. Football being the most popular game in the world, many studies have been conducted to analyze and predict match winners based on players’ physical and technical performance. In this study, we analyzed the matches from the professional football league of Qatar Stars League (QSL) covering the matches held in the last ten seasons. We incorporated the highest number of professional matches from the last ten seasons covering from 2011 up to 2022 and proposed SoccerNet, a Gated Recurrent Unit (GRU)-based deep learning-based model to predict match winners with over 80% accuracy. We considered match- and player-related information captured by STATS platform in a time slot of 15 minutes. Then we analyzed players’ performance at different positions on the field at different stages of the match. Our results indicated that in QSL, the defenders’ role in matches is more dominant than midfielders and forwarders. Moreover, our analysis suggests that the last 15–30 minutes of match segments of the matches from QSL have a more significant impact on the match result than other match segments. To the best of our knowledge, the proposed model is the first DL-based model in predicting match winners from any professional football leagues in the Middle East and North Africa (MENA) region. We believe the results will support the coaching staff and team management for QSL in designing game strategies and improve the overall quality of performance of the players.

In modern-day computing, cloud services are widely used in every aspect of life. So, user satisfaction depends on the effectiveness and efficiency of cloud services. Service broker policy of the cloud maintains the effectiveness and efficiency of cloud services. Service broker policy provides the rules and norms based on which a data center is selected for a userbase request. This paper proposes a genetic algorithm-based service broker policy that provides the optimal sequence of data centers for different userbases based on their requirements. This research aims to find an optimal data center for userbases that can achieve user satisfaction by minimizing the cloud service's response time and data processing time. We have experimented with our proposed genetic algorithm-based service broker policy in the CloudAnalyst platform based on different real-world scenarios. Simulation results indicate that our proposed genetic algorithm outperforms existing traditional algorithms. © 2023 IEEE.

Work in the area of Google Hacking resulted in two presentations, 2 publications, a grant of $2500, and incorporation of results in two courses (Web Security CSC553)and (Information Security CSC453).

Monitoring of electrical distribution grids requires the joint processing of electrical measurements from different grid locations. Such type of processing is influenced by inaccuracies in measurement data originating from measurement errors, non-ideal clocks in measurement devices, and from time averaging of measurands as part of the data collection process. This paper introduces an approach to assess the impact of these three different measurement artifacts in realistic measurement scenarios of electrical distribution grids. A case study of power loss calculation in a real-life medium-voltage grid is presented, covering both technical loss obtained from current measurement and total loss obtained from power measurements. The results show that total loss in general is more robust to aggregation of power measurements over longer measurement intervals, while it is more sensitive to measurement errors and clock offsets. The results of the study are important for quantifying the trustworthiness of the obtained loss values and for the future enhancement of the measurement data collection process. © 2023 ACM.

The optical disc in the human retina can reveal important information about a person's health and well-being. We propose a deep learning-based approach to automatically identify the region in human retinal images that corresponds to the optical disc. We formulated the task as an image segmentation problem that leverages multiple public-domain datasets of human retinal fundus images. Using an attention-based residual U-Net, we showed that the optical disc in a human retina image can be detected with more than 99% pixel-level accuracy and around 95% in Matthew's Correlation Coefficient. A comparison with variants of UNet with different encoder CNN architectures ascertains the superiority of the proposed approach across multiple metrics.

In this work, we propose a multi-task learning-based approach towards the localization of optic disc and fovea from human retinal fundus images using a deep learning-based approach. Formulating the task as an image-based regression problem, we propose a Densenet121-based architecture through an extensive set of experiments with a variety of CNN architectures. Our proposed approach achieved an average mean absolute error of only 13pixels (0.04%), mean squared error of 11 pixels (0.005%), and a root mean square error of only 0.02 (13%) on the IDRiD dataset.