Hybridisation of classifiers for anomaly detection in big data

In this article, a review and summarization of the emerging scientific approaches of deep learning (DL) on cybersecurity are provided, a structured and comprehensive overview of the various cyberattack detection methods is conducted, existing cyberattack detection methods based on DL is categorized. Methods covering attacks to deep learning based on generative adversarial networks (GAN) are investigated. The datasets used for the evaluation of the efficiency proposed by researchers for cyberattack detection methods are discussed. The statistical analysis of papers published on cybersecurity with the application of DL over the years is conducted. Existing commercial cybersecurity solutions developed on deep learning are described.

show abstract

“…Thehybridisationofclassifiersincreasestheperformanceofcyberattackdetectionsystems (Alguliyev, 2019).…”

Section: Hybrid Methodsmentioning

confidence: 99%

Deep Learning in Cybersecurity

İmamverdiyev

Abdullayeva

2020

International Journal of Cyber Warfare and Terrorism

View full text Add to dashboard Cite

show abstract

“…A tenant territory model was conceptualized and implemented at the initial layer, restoring the ability to manage the communication approach for VM service and maintaining security confinement of distinct tenant trade networks using Software Defined Networking (SDN). Alguliyev et al (2019) provide a semi-supervised classification strategy for identifying anomalies in cloud infrastructure performance metrics based on a combination of classifiers. The suggested approach for generating ensemble Naive Bayes uses the SMO, J48, IBK, multilayer perceptron as well as PART algorithms.…”

Section: Related Workmentioning

confidence: 99%

Cost-Efficient Deadline Constrained Scientific Workflow Scheduling in Infrastructure-as-a-Service Clouds by Disqualifying Tasks with Anomalies

Rahumath¹,

Rajendran²,

Mohanasundaram³

et al. 2022

Journal of Computer Science

View full text Add to dashboard Cite

Cloud computing has transformed the way businesses and consumers think about their data and businesses. As a result, cloud computing is described as the on-demand availability of all computer system resources via the Internet as a paid service. Enhancing security is a major problem in cloud computing, which is also a major research topic because data is stored and processed in remote locations held by third parties. Another key research subject is the allocation of virtual machines to incoming workloads to decrease the cost of consumers' workload execution in cloud environments. Both of the aforementioned difficulties are addressed in this study. Deadline-constrained workflows are submitted to the application server, which goes through a pre-processing step, that identifies the presence of anomalies in the workflow tasks and disqualifies those tasks with anomalies, and schedules the adjusted workloads into heterogeneous Virtual machines using a modified-PCP algorithm. Our approach is compared to the IC-LOSS and IC-PCP algorithms that are already in use. In comparison to the existing IC-PCP and IC-LOSS algorithms, experimental results suggest that using the modified-PCP algorithm for deadline constrained workflows after deleting those anomalous tasks produces better results.

show abstract

“…Most of the researchers study static software defect prediction techniques and, in this paper, also the static software prediction techniques are mainly studied [4]. e study of static software defect prediction techniques is divided into three main aspects: first, how to evaluate software defect prediction models; second, for the problem of choosing software metrics, effectively choosing metrics applicable to software defect prediction; and third, which qualitative or quantitative or hybrid models can be applied to software defect prediction [2,13]. Regarding evaluation metrics, precision, clarity, and sensitivity are frequently used evaluation metrics.…”

Section: Current Status Of Researchmentioning

confidence: 99%

“…Although the collected data is concentrated, most of the malicious samples are 64-bit programs, but there are still 417 32-bit programs. However, with the development of hardware technology and the popularity of computers, the scale and complexity of software became larger and larger, and the previous way of software development became increasingly difficult, and, to solve the resulting "software crisis," the way of software development was gradually systematized and engineered [2]. However, in the software development process, the existence of software defects is inevitable due to the limitations of resources and developers' experience.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Software Defect Prediction Methods for Cloud Environments Research

Liu

Wang

2021

Scientific Programming

View full text Add to dashboard Cite

This paper provides an in-depth study and analysis of software defect prediction methods in a cloud environment and uses a deep learning approach to justify software prediction. A cost penalty term is added to the supervised part of the deep ladder network; that is, the misclassification cost of different classes is added to the model. A cost-sensitive deep ladder network-based software defect prediction model is proposed, which effectively mitigates the negative impact of the class imbalance problem on defect prediction. To address the problem of lack or insufficiency of historical data from the same project, a flow learning-based geodesic cross-project software defect prediction method is proposed. Drawing on data information from other projects, a migration learning approach was used to embed the source and target datasets into a Gaussian manifold. The kernel encapsulates the incremental changes between the differences and commonalities between the two domains. To this point, the subspace is the space of two distributional approximations formed by the source and target data transformations, with traditional in-project software defect classifiers used to predict labels. It is found that real-time defect prediction is more practical because it has a smaller amount of code to review; only individual changes need to be reviewed rather than entire files or packages while making it easier for developers to assign fixes to defects. More importantly, this paper combines deep belief network techniques with real-time defect prediction at a fine-grained level and TCA techniques to deal with data imbalance and proposes an improved deep belief network approach for real-time defect prediction, while trying to change the machine learning classifier underlying DBN for different experimental studies, and the results not only validate the effectiveness of using TCA techniques to solve the data imbalance problem but also show that the defect prediction model learned by the improved method in this paper has better prediction performance.

show abstract

Hybridisation of classifiers for anomaly detection in big data

Cited by 7 publications

References 21 publications

Deep Learning in Cybersecurity

Deep Learning in Cybersecurity

Cost-Efficient Deadline Constrained Scientific Workflow Scheduling in Infrastructure-as-a-Service Clouds by Disqualifying Tasks with Anomalies

Deep Learning Software Defect Prediction Methods for Cloud Environments Research

Contact Info

Product

Resources

About