Abstract Denial of Service (DoS) and Distributed Denial of Service (DDoS) attacks are among the most lethal cyber threats in this world, which make an online service unavailable to its legitimate users by overwhelming the service provider’s resources. Regarding the importance of online services in a human's life, researchers have been working on techniques to detect and mitigate these kinds of attacks. Machine-learning models showed acceptable performance in DDoS detection. Hence, in this paper, we developed a machine-learning model for classifying network traffic and detecting DDoS attacks using a unique approach to pre-process the data. The most innovative aspect of our work is deploying our developed machine-learning model into an online real-time DDoS detection system and testing it under real DDoS attacks. Implementing and testing a DDoS detection system that can work outside of a dataset and can be used against real threats was the missing part of other similar works that were done in this paper. The model on offline data and the system under real attacks both showed great performance in detecting attack traffic with accuracies of 99.99% and 95.30%, respectively, and proved they can effectively be used against DDoS attacks.

Breast cancer is a leading public health concern that demands improved strategies for early diagnosis and prognosis. In this study, we propose P3DE, a novel computational framework that combines deep ensemble learning with the Parameter-less Population Pyramid (P3) metaheuristic for the identification of breast cancer biomarkers from gene expression data. P3DE integrates multiple autoencoders with diverse activation functions and dynamically computes ensemble weights based on reconstruction performance, enabling robust and adaptive feature selection without manual parameter tuning. Applied to the GSE42568 dataset, P3DE achieved outstanding classification performance (97.22% accuracy, 100% precision, 96.88% recall, 98.41% F1 and F2 scores, and 94.02% TMCC), outperforming conventional and state-of-the-art metaheuristic methods. Biological analysis revealed that the selected genes are enriched in key pathways and processes associated with breast cancer, such as peptide hormone response, membrane raft signaling, protein heterodimerization, and the PPAR signaling pathway. Notably, several identified genes (ALDH1A1, FGF2, TOP2A, EPCAM, CDH1) are linked to drugs, emphasizing their potential in targeted therapy. These results demonstrate the effectiveness of P3DE in uncovering biologically meaningful and clinically actionable biomarkers, highlighting the promise of hybrid, parameter-free computational models in precision oncology.

Breast cancer remains a global health challenge, and early detection is crucial for improving survival rates. However, traditional biomarker detection methods in machine learning face challenges such as high false positives, large gene datasets, and limited sample sizes. These challenges are exacerbated by the limitations of traditional differential evolution techniques, which struggle with scalability and effectiveness in high-dimensional, complex problems. This study introduces Ensemble-Based Logical Binary QANA (LBQANA_En), an improved differential evolution variant tailored for large-scale global optimization in gene expression analysis. By integrating six gene expression datasets, LBQANA_En overcomes the constraints of small sample sizes and the complexities inherent in gene expression data. Inspired by the quantum navigation of migratory birds and leveraging logical operators like XOR and OR, LBQANA_En outperforms other binary QANA variants in biomarker detection. The algorithm successfully identifies key biomarkers, including LPL, LEP, CD36, CDC20, TOP2A, and EZH2, significantly improving breast cancer detection accuracy and reducing false positives, achieving an impressive F1 score of 98.958%. These biomarkers provide critical insights into important pathways, such as AMPK and PPAR signaling, setting a new benchmark in computational biology and bioinformatics research, and paving the way for advancements in diagnostic techniques and medical science.

Breast cancer remains a major public health challenge worldwide. The identification of accurate biomarkers is critical for the early detection and effective treatment of breast cancer. This study utilizes an integrative machine learning approach to analyze breast cancer gene expression data for superior biomarker and drug target discovery. Gene expression datasets, obtained from the GEO database, were merged post-preprocessing. From the merged dataset, differential expression analysis between breast cancer and normal samples revealed 164 differentially expressed genes. Meanwhile, a separate gene expression dataset revealed 350 differentially expressed genes. Additionally, the BGWO_SA_Ens algorithm, integrating binary grey wolf optimization and simulated annealing with an ensemble classifier, was employed on gene expression datasets to identify predictive genes including TOP2A, AKR1C3, EZH2, MMP1, EDNRB, S100B, and SPP1. From over 10,000 genes, BGWO_SA_Ens identified 1404 in the merged dataset (F1 score: 0.981, PR-AUC: 0.998, ROC-AUC: 0.995) and 1710 in the GSE45827 dataset (F1 score: 0.965, PR-AUC: 0.986, ROC-AUC: 0.972). The intersection of DEGs and BGWO_SA_Ens selected genes revealed 35 superior genes that were consistently significant across methods. Enrichment analyses uncovered the involvement of these superior genes in key pathways such as AMPK, Adipocytokine, and PPAR signaling. Protein-protein interaction network analysis highlighted subnetworks and central nodes. Finally, a drug-gene interaction investigation revealed connections between superior genes and anticancer drugs. Collectively, the machine learning workflow identified a robust gene signature for breast cancer, illuminated their biological roles, interactions and therapeutic associations, and underscored the potential of computational approaches in biomarker discovery and precision oncology.

This study contributes to the literature by introducing the role of “location information sharing strategy” in optimizing the closed-loop supply chain (CLSC) performance. In the investigated CLSC, one manufacturer collects the used products through two collectors. The collectors may face collection disruptions and adopt thelocation information sharing strategyto overcome the disruptions. More precisely, the collector who faces disruption shares the customers’ location information with its rival to collect the used products. On the other hand, the manufacturer invests in the corporate social responsibility effort, which has bilateral effects on the forward and reverse flows. The products are sold through online and retail channels. The problem is first formulated in decentralized and centralized settings. Afterward, a coordination mechanism is developed to maximize profitability. The data from the case study is used to evaluate the performance of the proposed structures. The results indicate that the developed contract improves profitability and sustainability. Moreover, it is shown that thelocation information sharing strategycan be an efficient strategy for managers to overcome collection disruptions.

Abstract The existence of fraud in credit card transactions causes many financial losses leading to customers’ loss of trust. Fraud detection methods based on machine learning techniques prevent such losses. Despite the literature on fraud detection, there is a lack of algorithms that detect fraud with acceptable performance in the credit card fraud detection field. Therefore, this study proposed a comprehensive ensemble-based method using an efficient weighted voting strategy for credit card fraud detection that can address the previous algorithms’ weaknesses. First, since the dataset is imbalanced, the proposed method balanced the dataset by stratifying it into three different proportions of normal and fraudulent transactions (1 to 1, 1 to 4 and 1 to 9 ratios). The features in each dataset are ranked by four feature-ranking methods, and the Random Forest classifier is applied to each of them for selecting the essential features. Afterward, using the seven base classifiers and the chosen features, 12 ensembles have been developed. Besides, a weighted voting strategy is proposed, and the fraudulent transactions are detected through voting based on the base classifiers’ and ensembles’ weights, which are calculated by their performance. The computational results indicated that the suggested Eclf10 is the best ensemble and its Logistic Regression classifier also has the best performance among other base classifiers. The Eclf10 leads to 99.97% accuracy, 87.78% precision, 97.70% recall, 92.21% F1-score and 95.634% F2-score, which has a superiority over the previous ensemble-based methods (e.g. majority voting ensemble, stacking classifier, Adaboost, Gradient Boosting).

Among the novel IT paradigms, cloud computing and the Internet of Things (CloudIoT) are two complementary areas designed to support the creation of smart cities and application services. The CloudIoT not only presents ubiquitous services through IoT nodes but it also provides virtually unlimited resources through services composition. The services composition problem aims to find a set of services among functionally equivalent services with different Quality of Service (QoS) concerning users' constraints. To this aim, previous studies calculate QoS values through service logs without considering the presence of anomalies in the existing QoS values; however, the dynamicity of distributed service environments and communication networks in CloudIoT environments causes anomalies in the QoS values. Therefore, existing approaches fail to model QoS values accurately that leads to service-level agreement (SLA) violation and penalties for service broker. To address this challenge, we propose a scalable anomaly-aware approach (SAIoT) including two main components: the first component models QoS values based on a machine learning anomaly detection technique, to remove the existing abnormal QoS records, and the second component finds a near-optimal composition by using an effective and efficient metaheuristic algorithm. The experimental results based on real-world data sets show that our approach achieves 30.64% of the average improvement in the QoS value of a composite plan with equal or even less price compared to the previous works, such as information theory-based and advertised QoS-based methods.

Abstract The usage of social networks shows a growing trend in recent years. Due to a large number of online social networking users, there is a lot of data within these networks. Recently, advances in technology have made it possible to extract useful information about individuals and the interactions among them. In parallel, several methods and techniques were proposed to preserve the users’ privacy through the anonymization of social network graphs. In this regard, the utilization of the k-anonymity method, where k is the required threshold of structural anonymity, is among the most useful techniques. In this technique, the nodes are clustered together to form the super-nodes of size at least k. Our main idea in this paper is, initially, to optimize the clustering process in the k-anonymity method by means of the particle swarm optimization (PSO) algorithm in order to minimize the normalized structural information loss (NSIL), which is equal to maximizing 1-NSIL. Although the proposed PSO-based method shows a higher convergence rate than the previously introduced genetic algorithm (GA) method, it did not provide a lower NSIL value. Therefore, in order to achieve the NSIL value provided by GA optimization while preserving the high convergence rate obtained from the PSO algorithm, we present hybrid solutions based on the GA and PSO algorithms. Eventually, in order to achieve indistinguishable nodes, the edge generalization process is employed based on their relationships. The simulation results demonstrate the efficiency of the proposed model to balance the maximized 1-NSIL and the algorithm’s convergence rate.

Abstract Online social networks have a strong potential to be divided into a number of dense substructures, called communities. In such heterogeneous networks, the communities refer not only to dense parts of links but also to clusters present among other dimensions such as users' profiles, comments, and information flows. To find communities in these networks, researchers have developed a number of methods; however, to the best of the authors' knowledge, these methods are limited in taking only 2 dimensions into account, and they are also not able to give a sense of how users behave in their communities. To deal with these issues, this paper proposes a multiobjective optimization model in which a specific objective function has been used for each considered dimension in a given network. Because of the NP‐hardness of the studied problem, an efficient and effective multiobjective metaheuristic algorithm has been developed. By juxtaposing the nondominated solutions obtained, the proposed algorithm can demonstrate how users behave in their communities. To illustrate the effectiveness of the algorithm, a set of experiments with a comprehensive evaluation method is provided. The results show the superiority and the stability of the proposed algorithm.

Online Social Network (OSN) users generate massive amounts of information by their online interactions, by publishing profiles and posting content. Detection and analysis of the dense sub-structures of networks, called communities could facilitate a comprehensive understanding of OSNs. This present s the challenge of formulating appropriate means to evaluate and validate each detected community. Most researchers have tackled this issue by comparing results obtained from community detection algorithms with information on available social grouping as a ground-truth. However, social grouping does not guarantee formation or existence of an experienced sense of community, based on the community-oriented behavior patterns of its users. This study presents a new scoring function that targets the behavior of nodes in order to validate detected communities. Indeed, we employ this function as a Cluster Validity Index (CVI) for evaluating detected communities. Then, performance of the proposed CVI was compared with other known functions by ranking in terms of several goodness metrics, on a variety of homogeneous networks. This study also presents an enhanced version of the CVI to evaluate communities efficiently in heterogeneous networks. A number of experiments have been provided to demonstrate the effectiveness and reliability of the proposed CVI for heterogeneous networks.

Security is considered one of the most important concerns in e-payment systems. Although using mobile devices for e-payment is growing rapidly, mobile devices have limitations such as memory and battery limitations. Many security schemes have been proposed for e-payments, however most of them are too heavy regarding mobile devices since they use public key cryptography for security objectives. In this paper, we propose a mobile payment protocol for credit-debit card payment systems wherein customer needs to use public key encryption and decryption slightly.

Privacy preserving data mining (PPDM) algorithms attempt to reduce the injuries to privacy caused by malicious parties during the rule mining process. Usually, these algorithms are designed for the semi-honest model, where participants do not deviate from the protocol. However, in the real-world, malicious parties may attempt to obtain the secret values of other parties by probing attacks or collusion. In this study, the authors study how to preserve the privacy of participants in a collusion-free model of the frequent itemset mining process, where the protocol protects against probing attacks and collusion. The mining of frequent itemsets is the main step of association rule mining algorithms, and, in this study, the authors propose two privacy-preserving frequent itemset mining algorithms for both two-party and multi-party states in a collusion-free model for vertically partitioned (heterogeneous) data; in addition, a privacy measuring technique is proposed, which quantifies privacy based on the amount of disclosed sensitive information.

Purpose In traditional commerce, an auction is known as a mechanism of determining the value of a commodity that does not have a fixed price. Auctions are exciting and an increasing number of transactions are performed through e‐auctions. But most current auctions cannot address all the important security requirements. Usually, auction systems force bidders and sellers to trust the auctioneer and, on the other hand, do not provide anonymity for bidders and sellers. This paper aims to solve these problems by presenting an efficient anonymous secure auction schema (ASAS) without a fully trustworthy auctioneer. Design/methodology/approach The paper analyzes security properties and the complexity of previous works in auction security and then proposes a new ASAS that is more secure and efficient than previous works. Finally, security properties and the complexity of the new schema and previous works are compared with one another. Findings The proposed auction protocol does not force bidders and sellers to trust the auctioneer. In addition, it provides anonymity for both of them. Owing to these newly added features and high degree of security of ASAS, it is suggested that its use in high‐value auctions should require tighter security. Originality/value The paper proposes a new schema for electronic auctions that is secure and efficient and, in addition, does not force bidders and sellers to trust the auctioneer.

Purpose This paper aims to study an enterprise resource planning (ERP) software selection problem. The primary goal of this paper is to propose a two‐phase procedure to select an ERP vendor and a suitable ERP software. Design/methodology/approach In the first phase of the proposed method the preliminary actions – such as constructing a project team, collecting all possible information about ERP vendors and systems, and identifying the ERP system characteristics – are established. In the second phase, the authors present a modular approach to ERP vendor and software selection and propose a 0‐1 programming model to minimize total costs associated with procurement and integration expenditures. Findings The proposed approach and the model are considered to be more useful for small manufacturing enterprises (SMEs). Originality/value In using the model for analyzing the data about a real case study that is a commercial SME and based on obtained results, some parameter values of the model for all SMEs are suggested.