Swarm Intelligence Techniques Research Articles

An Adaptive Spiral Strategy Dung Beetle Optimization Algorithm: Research and Applications.

The Dung Beetle Optimization (DBO) algorithm, a well-established swarm intelligence technique, has shown considerable promise in solving complex engineering design challenges. However, it is hampered by limitations such as suboptimal population initialization, sluggish search speeds, and restricted global exploration capabilities. To overcome these shortcomings, we propose an enhanced version termed Adaptive Spiral Strategy Dung Beetle Optimization (ADBO). Key enhancements include the application of the Gaussian Chaos strategy for a more effective population initialization, the integration of the Whale Spiral Search Strategy inspired by the Whale Optimization Algorithm, and the introduction of an adaptive weight factor to improve search efficiency and enhance global exploration capabilities. These improvements collectively elevate the performance of the DBO algorithm, significantly enhancing its ability to address intricate real-world problems. We evaluate the ADBO algorithm against a suite of benchmark algorithms using the CEC2017 test functions, demonstrating its superiority. Furthermore, we validate its effectiveness through applications in diverse engineering domains such as robot manipulator design, triangular linkage problems, and unmanned aerial vehicle (UAV) path planning, highlighting its impact on improving UAV safety and energy efficiency.

Intelligent Routing Approaches Based on Ant Colony Optimization for Dynamic Internet of Things Network

Introduction: This paper introduces two meta-heuristic approaches utilizing Swarm Intelligence and ant colony optimization techniques. The strategy comprises applying smart routing technology to optimize a dynamic IoT network computed path. Method: The issue of route selection to achieve the target and critical factors such as network energy, left energy in each gadget, run out IoT nodes count has been explored. After rigorous iterations extending up to 1000, the simulation has yielded results for two distinctive routing approaches. The ABED (ACO- Breadth first search- Euclidean- Dynamic) and the ADED (ACODijkstra algorithm -Euclidean- Dynamic) have simulated and compared their network efficiencies using MATLAB. Results: At node 200, ABED exhibits a performance advantage over ADED of 1.6%. This efficiency differential between ABED and ADED expands to 2.9% at 300 nodes and further to 2.6% at 400 nodes. Furthermore, ABED showcases superior network stability in routing techniques compared to ADED. Specifically, ABED's routing technique achieves a more consistent network compared to ADED. Conclusion: In networks comprising 500 nodes, ABED surpasses ADED by 13.33% in the context of HND (Half Node Dead) and by 6.7% in the case of LND (Last Node Dead).

CPU-GPU Cooperative QoS Optimization of Personalized Digital Healthcare Using Machine Learning and Swarm Intelligence.

In recent decades, the rapid advances in information technology have promoted a widespread deployment of medical cyber-physical systems (MCPS), especially in the area of digital healthcare. In digital healthcare, medical edge devices empowered by CPU-GPU (Graphics Processing Unit) cooperative multiprocessor system-on-chips (MPSoCs) have a great potential in processing and managing the massive amounts of health-related data. However, most of the existing works on CPU-GPU cooperative MPSoCs cannot maintain a high-precision workload estimation since they simply leverage the worst-case execution cycles to pessimistically predict the workload of digital healthcare applications. Besides, they neglect the personalized requirements of individual healthcare applications and the lifetime reliability demands of heterogeneous CPU-GPU cores. As a result, the normal functions of medical edge devices and the quality-of-services (QoS) of digital healthcare applications are likely to suffer from underlying failures and degradation. In this paper, we explore CPU-GPU cooperative QoS optimization of personalized digital healthcare applications running on reliability guaranteed edge devices with the help of machine learning and swarm intelligence techniques. We first develop two novel predictors: one is a machine learning based predictor for application workload estimation, and the other is a feature-driven predictor for application QoS estimation. We then incorporate the two predictors into a swarm intelligent application scheduling scheme upon the cooperative dual-population evolutionary algorithm (c-DPEA) to find optimal application mapping and partitioning settings. Experimental results show that our solution not only augments the average QoS of whole digital healthcare applications by 15.7%, but also balances the QoS of individual digital healthcare applications by 64.3%.

A Comprehensive Analysis of Deep Learning and Swarm Intelligence Techniques to Enhance Vehicular Ad-hoc NETwork Performance

A Comprehensive Analysis of Deep Learning and Swarm Intelligence Techniques to Enhance Vehicular Ad-hoc NETwork Performance

A Review of Metaheuristic Optimization Techniques in Text Classification

Metaheuristic algorithms, inspired by natural phenomena and human-based strategies, offer versatile approaches to navigate diverse search spaces and adapt to dynamic environments. These algorithms, including evolutionary algorithms, swarm intelligence, bio-inspired methods, human-based approaches, and plant-inspired techniques, have found applications across diverse domains such as engineering, finance, healthcare, logistics, and telecommunications. In the text classification domain, metaheuristic techniques have emerged as powerful tools to enhance the accuracy, efficiency, and robustness of classification systems. By optimizing feature subsets, fine-tuning model parameters, and addressing challenges such as feature selection, dimensionality reduction, class imbalance, and noisy data, metaheuristic algorithms provide flexible solutions that adapt to various text datasets and tasks. This review paper comprehensively explores recent advancements in metaheuristic applications in text classification across six categories. From evolutionary-based methods to swarm-based approaches, bio-inspired techniques to physics/chemistry-based strategies, human-based methods to plant-based algorithms, researchers have leveraged diverse metaheuristic techniques to push the boundaries of text classification. Through a systematic analysis of recent research studies, this review provides insights into the strengths, limitations, and future directions of metaheuristic optimization in the context of text classification.

Cross-Language Translation and Comparative Study of English Literature Using Machine Translation Algorithms

Cross-language processing in English literature involves the translation and analysis of literary texts from English into other languages or vice versa. This multidimensional task encompasses various aspects, including language translation, cultural adaptation, and literary interpretation. Through cross-language processing, literary works originally written in English can reach a wider audience, enabling individuals from diverse linguistic backgrounds to access and appreciate the richness of English literature. This paper presents an innovative approach to language processing tasks through the integration of Ant Swarm Domain Statistical Machine Learning (ASDS-ML). Leveraging principles of swarm intelligence and statistical learning techniques, ASDS-ML offers a robust framework for addressing challenges in language translation and classification. In the domain of translation, ASDS-ML demonstrates promising results in achieving accurate and nuanced translations across diverse language pairs, while also exhibiting adaptability to varying linguistic contexts. Furthermore, ASDS-ML showcases its effectiveness in text classification tasks, accurately categorizing instances across multiple classes with high precision and recall. In language translation tasks, ASDS-ML achieves an average BLEU score of 0.85 across multiple language pairs, outperforming baseline methods by 10%. Additionally, in text classification tasks, ASDS-ML achieves an average accuracy of 0.92 across ten different classes, surpassing existing approaches by 5%.

BS-SCRM: a novel approach to secure wireless sensor networks via blockchain and swarm intelligence techniques

In this paper, we present a novel Secure Clustering Routing Method based on Blockchain and Swarm Intelligence (BS-SCRM) for Wireless Sensor Networks (WSNs), which serves as a cornerstone in the Internet of Things (IoT) infrastructure. Recognizing the limitations of existing clustering routing methods in addressing security threats, our approach integrates blockchain technology to fortify WSNs against vulnerabilities such as man-in-the-middle attacks. The proposed BS-SCRM method is structured in two phases: (1) an enhanced cluster head (CH) election utilizing an elite strategy-enhanced Whale Optimization Algorithm (WOA) that considers node energy and proximity to the base station, and (2) a secure data on-chain phase where blockchain comes into play, encrypting and validating cluster data to safeguard integrity and prevent tampering. We further tackle the challenge of implementing blockchain in resource-constrained WSNs by assigning distinct roles to devices, i.e., ordinary nodes with data viewing permissions and accounting nodes entrusted with both data viewing and consensus algorithm execution. Extensive simulations confirm that BS-SCRM not only improves clustering quality but also provides a more secure and energy-efficient routing solution compared to contemporary methods. More specifically, simulation results in different scenarios demonstrate that BS-SCRM enhances network lifetime by 24–73% compared to other clustering methods when facing attacks.

Enhancing CNN‐LSTM neural networks using jellyfish search algorithm for pandemic modeling

SummaryThis paper presents a comprehensive three‐step approach (CNN‐JSO‐LSTM) for predictive modeling using a pandemic such as COVID‐19 as a test case. Initially, a Convolutional Neural Network (CNN) is employed to extract crucial features pertinent to the pandemic. Subsequently, the Jellyfish Search Optimizer (JSO) algorithm is applied for feature selection, identifying the most relevant factors. These chosen features are then inputted into a Long Short‐Term Memory (LSTM) network, responsible for classifying samples into “healthy” and “diseased” categories. Our method enhances LSTM performance using the Jellyfish Search optimizer, resulting in exceptional prediction accuracy. Our experiments achieved remarkable metrics, with an accuracy of 95.32%, high sensitivity (94.87%), and precision (94.28%), surpassing alternative methods. In conclusion, our study presents a promising and highly accurate approach for pandemic prediction, harnessing deep learning and swarm intelligence techniques. These findings suggest a potential for more effective pandemic management and intervention strategies.

Enhancing personalized learning with explainable AI: A chaotic particle swarm optimization based decision support system

In the realm of Educational Technology, personalized learning is pivotal, yet predicting students' learning abilities based on learning styles and ICT remains challenging. We propose a decision support system using Machine Learning (ML), swarm intelligence, and explainable artificial (XAI) techniques to assess students' performance. Our model employs Chaotic Particle Swarm Optimization (C-PSO) with Henon execution, outperforming Genetic Algorithm (GA), Ant Colony Optimization (ACO), Firefly Algorithm (FA), Bee Colony Optimization (BCO), Artificial Fish Swarm Algorithm (AFSA), Mayfly Optimization Algorithm (MFOA), Mother Optimization Algorithm (MOA), Fuzzy Self-Tuning PSO (FST-PSO). Evaluating efficiency, effectiveness, and solution quality reveals C-PSO's superiority. The study identifies the significant impact of ICT on self-progress and employs Spearman Rank correlation for statistical validation. Findings suggest C-PSO as an effective tool for optimizing educational data analysis and decision-making. Further exploration in real-world educational settings and comparative analyses with alternative optimization techniques are recommended for future research.

An efficient energy management scheme using rule-based swarm intelligence approach to support pulsed load via solar-powered battery-ultracapacitor hybrid energy system

This work presents an energy management scheme (EMS) based on a rule-based grasshopper optimization algorithm (RB-GOA) for a solar-powered battery-ultracapacitor hybrid system. The main objective is to efficiently meet pulsed load (PL) demands and extract maximum energy from the photovoltaic (PV) array. The proposed approach establishes a simple IF-THEN set of rules to define the search space, including PV, battery bank (BB), and ultracapacitor (UC) constraints. GOA then dynamically allocates power shares among PV, BB, and UC to meet PL demand based on these rules and search space. A comprehensive study is conducted to evaluate and compare the performance of the proposed technique with other well-known swarm intelligence techniques (SITs) such as the cuckoo search algorithm (CSA), gray wolf optimization (GWO), and salp swarm algorithm (SSA). Evaluation is carried out for various cases, including PV alone without any energy storage device, variable PV with a constant load, variable PV with PL cases, and PV with maximum power point tracking (MPPT). Comparative analysis shows that the proposed technique outperforms the other SITs in terms of reducing power surges caused by PV power or load transition, oscillation mitigation, and MPP tracking. Specifically, for the variable PV with constant load case, it reduces the power surge by 26%, 22%, and 8% compared to CSA, GWO, and SSA, respectively. It also mitigates oscillations twice as fast as CSA and GWO and more than three times as fast as SSA. Moreover, it reduces the power surge by 9 times compared to CSA and GWO and by 6 times compared to SSA in variable PV with the PL case. Furthermore, its MPP tracking speed is approximately 29% to 61% faster than its counterparts, regardless of weather conditions. The results demonstrate that the proposed EMS is superior to other SITs in keeping a stable output across PL demand, reducing power surges, and minimizing oscillations while maximizing the usage of PV energy.

Enhancing feature selection with a novel hybrid approach incorporating genetic algorithms and swarm intelligence techniques

Computing advances in data storage are leading to rapid growth in large-scale datasets. Using all features increases temporal/spatial complexity and negatively influences performance. Feature selection is a fundamental stage in data preprocessing, removing redundant and irrelevant features to minimize the number of features and enhance the performance of classification accuracy. Numerous optimization algorithms were employed to handle feature selection (FS) problems, and they outperform conventional FS techniques. However, there is no metaheuristic FS method that outperforms other optimization algorithms in many datasets. This motivated our study to incorporate the advantages of various optimization techniques to obtain a powerful technique that outperforms other methods in many datasets from different domains. In this article, a novel combined method GASI is developed using swarm intelligence (SI) based feature selection techniques and genetic algorithms (GA) that uses a multi-objective fitness function to seek the optimal subset of features. To assess the performance of the proposed approach, seven datasets have been collected from the UCI repository and exploited to test the newly established feature selection technique. The experimental results demonstrate that the suggested method GASI outperforms many powerful SI-based feature selection techniques studied. GASI obtains a better average fitness value and improves classification performance.

An automatic diagnostic model for the detection and classification of cardiovascular diseases based on swarm intelligence technique

Globally, cardiovascular diseases (CVDs) rank among the leading causes of mortality. One out of every three deaths is attributed to cardiovascular disease, according to new World Heart Federation research. Cardiovascular disease can be caused by a number of factors, including stress, alcohol, smoking, a poor diet, inactivity, and other medical disorders like high blood pressure or diabetes. In contrast, for the vast majority of heart disorders, early diagnosis of associated ailments results in permanent recovery. Using newly developed data analysis technology, examining a patient's medical record could aid in the early detection of cardiovascular disease. Recent work has employed machine learning algorithms to predict cardiovascular illness on clinical datasets. However, because of their enormous dimension and class imbalance, clinical datasets present serious issues. An inventive model is offered in this work for addressing these problems.An efficient decision support system, also known as an assistive system, is proposed in this paper for the diagnosis and classification of cardiovascular disorders. It makes use of an optimisation technique and a deep learning classifier. The efficacy of traditional techniques for predicting cardiovascular disease using medical data is anticipated to advance with the combination of the two methodologies. Deep learning systems can reduce mortality rates by predicting cardiovascular illness based on clinical data and the patient's severity level. For an adequate sample size of synthesized samples, the optimisation process chooses the right parameters to yield the best prediction from an enhanced classifier. The 99.58% accuracy was obtained by the proposed method. Also, PSNR, sensitivity, specificity, and other metrics were calculated in this work and compared with systems that are currently in use.

Advances and Challenges in Feature Selection Methods: A Comprehensive Review

The feature selection area in data analytics is explored through a comprehensive literature review, and the increasing areas that have a data dependency problem and are being resolved with feature selection are highlighted. Review topics of this course cover the foundations to present use cases, for example, cybersecurity, healthcare, and finance. Particularly crucially for the healthcare domain, it reduces the dimensionality and elucidates complex causal links. The further investigation overlaps contemporary techniques, including optimization-based methods, swarm intelligence and algorithms for the diagnosis of heart diseases. The conclusion builds on the practical assessment and underlines research gaps, serving as a basis to set a diversified technological review. This also exhibits new techniques that have released their efficiency in classification environments, for example, hybrid Ant Colony Optimization and the Gray Wolf Optimizer. The ISSA algorithm stands out as a swarm intelligence technique that is best among others. The paper concludes by demonstrating that feature selection goes beyond the preprocessing stage, but it instead stands as a vital part of the fields of machine learning and data science and thus aids the researchers in both retrospective analysis and forthcoming projects.

Exploring the energy landscape of graphynes for noble gas adsorption using swarm intelligence

Gas adsorption on one-atom-thick membranes is a growing technology for separation applications owing to its excellent energy efficiency. Herein, we investigate the adsorption of the noble gases, Ne, Ar and Kr on graphynes (GYs), a novel class of one-atom-thick carbon membranes using a swarm intelligence technique, namely particle swarm optimization (PSO). Modeling the adsorption of noble gas clusters on two-dimensional substrates requires a thorough examination of the energy landscape. The high dimensionality of the problem makes it tricky to employ ab initio methods for such studies, necessitating the use of a metaheuristic global optimization technique such as PSO. We explored the adsorption of 1–30 atoms of Ne, Ar and Kr on α-, β-, γ- and rhombic-GYs to predict the most suitable form of GY for the adsorption of each of the gases. Employing the dispersion-corrected density functional theory (DFT-D) data for the adsorption of single gas atoms as the reference data, we parametrized two empirical pairwise potentials, namely, Lennard-Jones (LJ) and improved Lennard-Jones (ILJ) potentials. We then analyzed the growth pattern as well as the energetics of adsorption using the parametrized potentials, in combination with the PSO technique, which enabled us to predict the best possible membrane for the adsorption of the noble gases: α-GY for Ne and γ-GY for Ar and Kr. The accuracy of our modeling approach is further validated against DFT-D computations thereby establishing that PSO, when combined with the ILJ potential, can serve as a computationally feasible approach for modeling gas adsorption on GYs.

Swarm intelligence technique for solving optimal power flow problem in the Nigeria power system

This paper deals on the implementation of the particle swarm optimization method for solving power flow problems in the Nigeria power system. The difficulty of solving optimal power flow (OPF) problems increases significantly with increasing network size and complexity. Some of the weakness of the conventional methods include: limited ability in solving real-world large scale optimization problems, weakness in handling constraints, poor convergence and slow computational time. The results gotten by implementing the Particle Swarm Optimization (PSO) algorithm in Nigeria 330kV 52-Bus network show that the total active power and reactive power losses were substantially reduced to 0.03MW and 0.05MVAR. The system time of convergence was faster at 0.5seconds. Also, the maximum line active and reactive power losses on line 1 to 2 were reduced greatly to 0.00MW and 0.01MVAR respectively. These show that the system was better optimized with the PSO algorithm than the conventional method.

Load balancing model for cloud environment using swarm intelligence technique

A distributed system with a shared resource pool offers cloud computing services. According to the provider’s policy, customers can enjoy continuous access to these resources. Every time a job is transferred to the cloud to be carried out, the environment must be appropriately planned. A sufficient number of virtual machines (VM) must be accessible on the backend to do this. As a result, the scheduling method determines how well the system functions. An intelligent scheduling algorithm distributes the jobs among all VMs to balance the overall workload. This problem falls into the category of NP-Hard problems and is regarded as a load balancing problem. With spider monkey optimization, we have implemented a fresh strategy for more dependable and efficient load balancing in cloud environments. The suggested optimization strategy aims to boost performance by choosing the least-loaded VM to distribute the workloads. The simulation results clearly show that the proposed algorithm performs better regarding load balancing, reaction time, make span and resource utilization. The experimental results outperform the available approaches.

Underwater image contrast enhancement through an intensity-randomised approach incorporating a swarm intelligence technique with unsupervised dual-step fusion

ABSTRACT Analysing objects in an underwater medium is difficult due to light problems in such an environment. Low contrast and visibility in water medium result in restricted information extraction. Some previous methods inadequately reduce underwater colour cast and improve the image contrast minimally. This work proposes intensity-randomised underwater image contrast enhancement (IRUCE), an improved enhancement method that integrates an unsupervised dual-step fusion technique to reduce blue-green colour cast, improve contrast, and enhance image brightness, especially in turbid and deep underwater media. IRUCE enhances the inferior colour channels and moderates the dominant colour channel to reduce the colour cast before the intensity-randomised approach is implemented. The intensity-randomised approach is designed to effectively distribute image intensity across all intensity levels. Next, a swarm intelligence algorithm is fused to perform median equalisation on all colour channels. The median intensity values of inferior colour channels are shifted towards the dominant colour channel. The unsharp masking technique is employed in the last step to increase image sharpness. The effectiveness of this approach is validated through quantitative and qualitative evaluations, and the results are compared with those of other state-of-the-art methods. Outcomes indicate that the proposed method improves underwater image quality significantly and outperforms other contemporary techniques.

A Deep Learning Framework with an Intermediate Layer Using the Swarm Intelligence Optimizer for Diagnosing Oral Squamous Cell Carcinoma.

One of the most prevalent cancers is oral squamous cell carcinoma, and preventing mortality from this disease primarily depends on early detection. Clinicians will greatly benefit from automated diagnostic techniques that analyze a patient's histopathology images to identify abnormal oral lesions. A deep learning framework was designed with an intermediate layer between feature extraction layers and classification layers for classifying the histopathological images into two categories, namely, normal and oral squamous cell carcinoma. The intermediate layer is constructed using the proposed swarm intelligence technique called the Modified Gorilla Troops Optimizer. While there are many optimization algorithms used in the literature for feature selection, weight updating, and optimal parameter identification in deep learning models, this work focuses on using optimization algorithms as an intermediate layer to convert extracted features into features that are better suited for classification. Three datasets comprising 2784 normal and 3632 oral squamous cell carcinoma subjects are considered in this work. Three popular CNN architectures, namely, InceptionV2, MobileNetV3, and EfficientNetB3, are investigated as feature extraction layers. Two fully connected Neural Network layers, batch normalization, and dropout are used as classification layers. With the best accuracy of 0.89 among the examined feature extraction models, MobileNetV3 exhibits good performance. This accuracy is increased to 0.95 when the suggested Modified Gorilla Troops Optimizer is used as an intermediary layer.

Exploring Hydrogen Storage Capacity in Metal-Organic Frameworks: A Bayesian Optimization Approach.

Metal-organic Frameworks (MOFs) can be employed for gas storage, capture, and sensing. Finding the MOF with the best adsorption property from a big database is usual for adsorption calculations. In high-throughput computational research, the expense of computing thermodynamic quantities limits the finding of MOFs for separations and storage.In this work, we demonstrate the usefulness of Bayesian optimization (BO) for estimating the H2 uptake capability of MOFs by using an existing dataset containing 98 000 real as well as hypothetical MOFs. It is demonstrated that in order to recover the best candidate MOFs, less than 0.027% of the database needs to be screened using the BO method. This allows future adsorption experiments on a small sample of MOFs to be undertaken with minimal experimental effort by effectively screening MOF databases. In addition, the presented BO can provide comprehensible material design insights, and the framework will be transferable to the setting of other target properties. We suggest using particle swarm optimisation (PSO), a swarm intelligence technique in artificial intelligence, to estimate MOFs' H2 uptake potential to achieve results comparable to BO. In addition, we have implemented the novel evolutionary-PSO (EPSO) method for comparison research and found some extremely interesting outcomes.

SYNTONY: Potential-aware fuzzing with particle swarm optimization

Fuzzing has gained significant traction in academic research as well as industry thanks to its effectiveness for discovering software vulnerabilities. However, even the state-of-the-art fuzzers are not very efficient at identifying promising seeds. Coverage-guided fuzzers, while fast and scalable, usually employ single criterion to evaluate the quality of seeds, which may incur bias and pass up optimal seeds. In this paper, we devise a novel potential-aware fuzzing scheme, namely SYNTONY, which seeks to measure seed potential utilizing multiple objectives and prioritize promising seeds that are more likely to generate interesting seeds via mutation. More specifically, SYNTONY leverages efficient swarm intelligence techniques like Particle Swarm Optimization (PSO) to explore multi-criteria seed selection, which allows SYNTONY to choose effectively promising seeds. Furthermore, we introduce decent power scheduling strategy to discover significantly more paths or crashes by gravitating towards more potential seeds. We implement this scheme on top of several state-of-the-art fuzzers, i.e., AFL, AFL++, FairFuzz, and PTFuzz. Our evaluations on 11 popular real-world programs demonstrate that SYNTONY significantly increases the number of unique crashes triggered and edge coverage discovered by 132.06 % and 28.69 % over AFL++. Further comparison also shows that SYNTONY outperforms other state-of-the-art fuzzers, e.g., AFL, FairFuzz, and PTFuzz. Also, extensive evaluations illustrate that SYNTONY provides a great compatibility and expansibility, while introducing negligible overhead.