IoT Bonet and Network Intrusion Detection using Dimensionality Reduction and Supervised Machine Learning

Desai, Madhuri Gurunathrao; Shi, Yong; Suo, Kun

doi:10.1109/uemcon51285.2020.9298146

Cited by 10 publications

(8 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, in contrast to the flow-based approach, instead of generating an aggregate statistic for each flow, we utilise the changes in the network data that occur as each packet arrives. While similar approaches have been used in the literature [48]- [50], we present a different approach in terms of both methodology and features. This uses both rolling (RW) and expanding windows (EW) that extract features from information carried by packets between the source and destination (MAC/IP address).…”

Section: Feature Extraction Methodsmentioning

confidence: 99%

“…The common characteristics of these tools are that all of the generated features are flow-based. Unlike these tools, Kitsune, used in [47]- [50] uses a sliding window approach, converting pcap files into a dataset with 115 features.…”

Section: B Featuresmentioning

confidence: 99%

“…Researchers have often evaluated their approach using more than one ML model. In many studies with multiple comparisons, ensemble models (such as RF and XGB) gave the best results [11]- [13], [16], [19]- [22], [25]- [29], [48]- [50], [55], [56].…”

Section: Machine Learning and Evaluation Metricsmentioning

confidence: 99%

“…For example, the IoTID20 [54] dataset was created using CICFlowMeter from IoT-NID pcap files. So, both flow [54] and sliding window [50] features were extracted from the IoT-NID dataset. On the other hand, KDD99, NSL-KDD, CIDDS-01, DS2OS, MQTTset, N-BaIoT and InSDN datasets do not include pcap files.…”

Section: B Featuresmentioning

confidence: 99%

See 3 more Smart Citations

IoTDevID: A Behavior-Based Device Identification Method for the IoT

Kahraman

Just

Lones

2022

IEEE Internet Things J.

View full text Add to dashboard Cite

While the use of the Internet of Things is becoming more and more popular, many security vulnerabilities are emerging with the large number of devices being introduced to the market. In this environment, IoT device identification methods provide a preventive security measure as an important factor in identifying these devices and detecting the vulnerabilities they suffer from. In this study, we present a method that identifies devices in the Aalto dataset using the convolutional neural network (CNN).

show abstract

Section: Feature Extraction Methodsmentioning

confidence: 99%

Section: B Featuresmentioning

confidence: 99%

Section: Machine Learning and Evaluation Metricsmentioning

confidence: 99%

Section: B Featuresmentioning

confidence: 99%

See 2 more Smart Citations

IoTDevID: A Behavior-Based Device Identification Method for the IoT

Kahraman

Just

Lones

2022

IEEE Internet Things J.

View full text Add to dashboard Cite

show abstract

“…Desai et al presented, in 2020, an intrusion and botnet detection system for IoT devices [ 11 ]. The proposed system built a multiclass-classifier using supervised learning models with Principal Component Analysis (PCA) for dimensionality reduction.…”

Section: Related Workmentioning

confidence: 99%

Towards an Explainable Universal Feature Set for IoT Intrusion Detection

Alani

Miri

2022

Sensors

View full text Add to dashboard Cite

As IoT devices’ adoption grows rapidly, security plays an important role in our daily lives. As part of the effort to counter these security threats in recent years, many IoT intrusion detection datasets were presented, such as TON_IoT, BoT-IoT, and Aposemat IoT-23. These datasets were used to build many machine learning-based IoT intrusion detection models. In this research, we present an explainable and efficient method for selecting the most effective universal features from IoT intrusion detection datasets that can help in producing highly-accurate and efficient machine learning-based intrusion detection systems. The proposed method was applied to TON_IoT, Aposemat IoT-23, and IoT-ID datasets and resulted in the selection of six universal network-flow features. The proposed method was tested and produced a high accuracy of 99.62% with a prediction time reduced by up to 70%. To provide better insight into the operation of the classifier, a Shapley additive explanation was used to explain the selected features and to prove the alignment of the explanation with current attack techniques.

show abstract