Applications of Machine Learning in Networking: A Survey of Current Issues and Future Challenges

Ridwan, M. A.; Radzi, Nurul Asyikin Mohamed; Abdullah, Fairuz; Jalil, Yanti Erana

doi:10.1109/access.2021.3069210

Cited by 25 publications

(17 citation statements)

References 124 publications

(189 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Machine learning techniques [8] are widely used for building Intrusion Detection Systems. In this context, classification refers to the process of using machine learning algorithms to identify normal versus malicious activity within a dataset, representing network traffic, for designing an anomaly-based IDS.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

A novel time efficient learning-based approach for smart intrusion detection system

2021

View full text Add to dashboard Cite

Background The ever increasing sophistication of intrusion approaches has led to the dire necessity for developing Intrusion Detection Systems with optimal efficacy. However, existing Intrusion Detection Systems have been developed using outdated attack datasets, with more focus on prediction accuracy and less on prediction latency. The smart Intrusion Detection System framework evolution looks forward to designing and deploying security systems that use various parameters for analyzing current and dynamic traffic trends and are highly time-efficient in predicting intrusions. Aims This paper proposes a novel approach for a time-efficient and smart Intrusion Detection System. Method Herein, we propose a Hybrid Feature Selection approach that aims to reduce the prediction latency without affecting attack prediction performance by lowering the model's complexity. Light Gradient Boosting Machine (LightGBM), a fast gradient boosting framework, is used to build the model on the latest CIC-IDS 2018 dataset. Results The proposed feature selection reduces the prediction latency ranging from 44.52% to 2.25% and the model building time ranging from 52.68% to 17.94% in various algorithms on the CIC-IDS 2018 dataset. The proposed model with hybrid feature selection and LightGBM gives 97.73% accuracy, 96% sensitivity, 99.3% precision rate, and comparatively low prediction latency. The proposed model successfully achieved a raise of 1.5% in accuracy rate and 3% precision rate over the existing model. An in-depth analysis of network parameters is also performed, which gives a deep insight into the variation of network parameters during the benign and malicious sessions.

show abstract

Section: Related Workmentioning

confidence: 99%

“…These metrics are calculated with the following Eqs. (5,6,7,8,9) using different negative and positive cases as TP-True positive, TN-True Negative, FP-False Positive, N-alse Negative.…”

Section: Evaluation Metricsmentioning

confidence: 99%

A novel time efficient learning-based approach for smart intrusion detection system

2021

View full text Add to dashboard Cite

show abstract

“…Our background in algorithms as opposed to networks gives us a different perspective on the survey. A recent survey was published by Ridwan et al in [64]. The authors presented a survey of recent trends in the application of ML in networks.…”

Section: Overview Of ML Applications In Networkmentioning

confidence: 99%

A Survey of Machine Learning Techniques for Video Quality Prediction from Quality of Delivery Metrics

2021

View full text Add to dashboard Cite

A growing number of video streaming networks are incorporating machine learning (ML) applications. The growth of video streaming services places enormous pressure on network and video content providers who need to proactively maintain high levels of video quality. ML has been applied to predict the quality of video streams. Quality of delivery (QoD) measurements, which capture the end-to-end performances of network services, have been leveraged in video quality prediction. The drive for end-to-end encryption, for privacy and digital rights management, has brought about a lack of visibility for operators who desire insights from video quality metrics. In response, numerous solutions have been proposed to tackle the challenge of video quality prediction from QoD-derived metrics. This survey provides a review of studies that focus on ML techniques for predicting the QoD metrics in video streaming services. In the context of video quality measurements, we focus on QoD metrics, which are not tied to a particular type of video streaming service. Unlike previous reviews in the area, this contribution considers papers published between 2016 and 2021. Approaches for predicting QoD for video are grouped under the following headings: (1) video quality prediction under QoD impairments, (2) prediction of video quality from encrypted video streaming traffic, (3) predicting the video quality in HAS applications, (4) predicting the video quality in SDN applications, (5) predicting the video quality in wireless settings, and (6) predicting the video quality in WebRTC applications. Throughout the survey, some research challenges and directions in this area are discussed, including (1) machine learning over deep learning; (2) adaptive deep learning for improved video delivery; (3) computational cost and interpretability; (4) self-healing networks and failure recovery. The survey findings reveal that traditional ML algorithms are the most widely adopted models for solving video quality prediction problems. This family of algorithms has a lot of potential because they are well understood, easy to deploy, and have lower computational requirements than deep learning techniques.

show abstract

“…The firewall filters by comparing the offset time of BHP and the real delay between the BHP and the accompanying DB. In [2], the authors designed and implemented an algorithm to classify the ingress nodes of an OBS network into three classes, i.e., Trusted, Blocked, and Suspicious. Based on the node's behavior and the amount of unutilized reserved resources, the classification was performed.…”

Section: Literature Reviewmentioning

confidence: 99%

“…The authors suggested that using data mining algorithms, one can find the hidden relation or pattern of behavior related to network performance and network control parameters. In [2], M. A. Ridwan et al presented a detailed review on the current trends in the application of machine learning in communication networks. The authors efficiently surveyed the existing literatures published in the period between 2017 and 2020 and listed the significant works on the application of machine learning in vulnerability prediction, routing, Quality of Service enhancement, intrusion detection, resource management, etc.…”

Section: Introductionmentioning

confidence: 99%

A Comparative Analysis of Semi-Supervised Learning in Detecting Burst Header Packet Flooding Attack in Optical Burst Switching Network

2021

View full text Add to dashboard Cite

This paper presents a comparative analysis of four semi-supervised machine learning (SSML) algorithms for detecting malicious nodes in an optical burst switching (OBS) network. The SSML approaches include a modified version of K-means clustering, a Gaussian mixture model (GMM), a classical self-training (ST) model, and a modified version of self-training (MST) model. All the four approaches work in semi-supervised fashion, while the MST uses an ensemble of classifiers for the final decision making. SSML approaches are particularly useful when a limited number of labeled data is available for training and validation of the classification model. Manual labeling of a large dataset is complex and time consuming. It is even worse for the OBS network data. SSML can be used to leverage the unlabeled data for making a better prediction than using a smaller set of labelled data. We evaluated the performance of four SSML approaches for two (Behaving, Not-behaving), three (Behaving, Not-behaving, and Potentially Not-behaving), and four (No-Block, Block, NB- wait and NB-No-Block) class classifications using precision, recall, and F1 score. In case of the two-class classification, the K-means and GMM-based approaches performed better than the others. In case of the three-class classification, the K-means and the classical ST approaches performed better than the others. In case of the four-class classification, the MST showed the best performance. Finally, the SSML approaches were compared with two supervised learning (SL) based approaches. The comparison results showed that the SSML based approaches outperform when a smaller sized labeled data is available to train the classification models.

show abstract

Applications of Machine Learning in Networking: A Survey of Current Issues and Future Challenges

Cited by 25 publications

References 124 publications

A novel time efficient learning-based approach for smart intrusion detection system

A novel time efficient learning-based approach for smart intrusion detection system

A Survey of Machine Learning Techniques for Video Quality Prediction from Quality of Delivery Metrics

A Comparative Analysis of Semi-Supervised Learning in Detecting Burst Header Packet Flooding Attack in Optical Burst Switching Network

Contact Info

Product

Resources

About