A review on machine learning based intrusion detection system for internet of things enabled environment

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 14, No. 2, April 2024, pp. 1890~1898
ISSN: 2088-8708, DOI: 10.11591/ijece.v14i2.pp1890-1898  1890
Journal homepage: http://ijece.iaescore.com
A review on machine learning based intrusion detection system
for internet of things enabled environment
Nisha, Nasib Singh Gill, Preeti Gulia
Department of Computer Science and Applications, Maharshi Dayanand University, Rohtak, India
Article Info ABSTRACT
Article history:
Received Jul 2, 2023
Revised Oct 12, 2023
Accepted Nov 29, 2023
Within an internet of things (IoT) environment, the fundamental purpose of
various devices is to gather the abundant amount of data that is being
generated and then transmit this data to the predetermined server over the
internet. IoT connects billions of objects and the internet to communicate
without human intervention. But network security and privacy issues are
increasing very fast, in today's world. Because of the prevalence of
technological advancement in regular activities, internet security has evolved
into a necessary requirement. Because technology is integrated into every
aspect of contemporary life, cyberattacks on the internet of things represent a
bigger danger than attacks against traditional networks. Researchers have
found that combining machine learning techniques into an intrusion
detection system (IDS) is an efficient way to get beyond the limitations of
conventional IDSs in an IoT context. This research presents a
comprehensive literature assessment and develops an intrusion detection
system that makes use of machine learning techniques to address security
problems in an IoT environment. Along with a comprehensive look at the
state of the art in terms of intrusion detection systems for IoT-enabled
environments, this study also examines the attributes of approaches,
common datasets, and existing methods utilized to construct such systems.
Keywords:
Attacks in internet of things
Environment
Internet of things
Internet of things security
Intrusion detection system
Machine learning
This is an open access article under the CC BY-SA license.
Corresponding Author:
Nisha
Department of Computer Science and Applications, Maharshi Dayanand University
Rohtak, Haryana, India
Email: nisha.rs.dcsa@mdurohtak.ac.in
1. INTRODUCTION
With the advent of the internet of things (IoT), computers have been imagined as everyday objects
that can sense their environment, form relationships with one another, and share information with one
another and the rest of the world over the Internet. The fundamental goal of IoT is to increase productivity
and speed up the delivery of critical information by having self-reported machines in a real-time setting.
There have been significant advancements in user experience in recent years thanks to IoT-based frameworks
in fields as diverse as smart farming, intelligent healthcare system, supply chain management, smart housing,
and intelligent transportation system [1], [2]. IoT devices are typically embedded with sensors, which
monitor and control the information over the internet for making the best decisions. By 2025, there will be
41.6 billion IoT devices, and they will produce 79.4 zettabytes of data [3], according to the international data
corporation (IDC). This technological advancement provides enormous benefits to consumers, but the IoT
has recently faced significant challenges due to security issues caused by equipment failure and malicious
attacks led by external intruders. IoT networks are extremely susceptible to online threats like distributed
denial-of-service (DDoS) assaults [4]. The variety of devices as well as protocols, direct accessibility of
equipment to the Internet, as well as resource limits on devices, are all factors that make it difficult to protect

Int J Elec & Comp Eng ISSN: 2088-8708 
A review on machine learning based intrusion detection system for internet of things … (Nisha)
1891
internet of things devices from being attacker [5]. An intrusion occurs whenever there is an attempt to
compromise the security, privacy, or availability of a system's resources [6]. Therefore, a specialized
component is necessary for protecting IoT networks. Increasing the intrusion detection system (IDS)
capabilities of wireless networks can help protect the IoT network against attacks and other vulnerabilities
[7]. As a result, IDSs are required to detect attacks so that IoT networks can continue to be reliable and
accessible. As the primary function of IDS is to detect attacks, it is crucial to define the many threats that can
arise in an IoT environment. Sybil attacks, selective forwarding attacks, service assaults, wormhole attacks,
sinkhole attacks, fake data attacks, black hole attacks, jammer attacks, and so on are all examples of key
attacks in the internet of things. Figure 1 depicts many sorts of attacks in the IoT context.
Figure 1. Attacks in IoT environment
Wormhole attack (routing-based assault), as explained by Deshmukh-Bhosale and Sonavane [8] in
their study work, involves simultaneous attacks or penetrations from two different directions in order to reach
the intended node. This study effort is praised for detecting nearby networks and calculating the effects of
threats. Rathore and Park [9] focused on the effects of the Sybil attack, where the node generates many
identities, resulting in a variety of routing protocols. To detect such assaults, detection techniques and
spoofing mechanisms are needed. Lyu et al. [10] presented intrusion detection methodology to identify
hostile users on IoT networks. In particular, the focus was on denial-of-service attacks, in which the attacker
attempts to take control of a node by overwhelming its resources. Furthermore, because the Internet of
Things devices frequently perform in an unsupervised environment, an adversary with malicious intent can
get physical access to these devices [11]. As a result, cyber-attacks are more likely to affect Internet of
Things platforms than traditional computer networks.
2. MACHINE LEARNING APPROACHES FOR IDS
Machine learning's ultimate goal is to endow machines with the capacity for autonomous learning
and decision-making, on par with human beings. Algorithms trained with machine learning utilize statistical
methods to examine large data sets for recurring patterns, then use that information to inform future decisions
and forecasts [12], [13]. Here, we give a brief introduction to the various machine-learning strategies
currently being used by IDSs in the IoT landscape (IDSs). In Table 1, a concise review of machine learning
(ML) approaches, including their benefits and drawbacks, as well as references to relevant research work that
has been carried out, is provided. The most prevalent machine learning approaches that are utilized for
creating IDSs in IoT systems are shown in Figure 2.
It is possible to utilize a variety of techniques to protect the backend IoT networks. Supervised
learning is more successful when the environment variable is known (the outcome should be the same for
every input). In cases where the outcome is irrelevant, unsupervised learning is employed; this type of
learning is typically used to classify the attributes.

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 14, No. 2, April 2024: 1890-1898
1892
Figure 3 depicts the supervised learning algorithm, random forest. There is a one-to-one correlation
between the density of a forest and the reliability of the decisions made using different types of decision
trees. The algorithm's fundamental component is a decision tree, which also acts as a decision aid. Using a
tree-like graph, we may depict several outcomes. These methods can be used for both classification and
regression. In supervised learning, the support vector machine is a popular technique (SVM). It works well
with both categorization and statistical analysis. However, it is mainly applied to categorization issues. See
Figure 4 for an illustration of how this classification method works to locate the hyperplane that demarcates
the two groups. The value of a specific coordinator is used here to represent each data point.
Table 1. A taxonomy of ML-based security techniques for internet of things systems
ML
method
Attack types
handled
Pros Cons Results
KB
[14]
HTTP attacks
(buffer overflow,
shell attacks),
DoS, probe, R2L
The training just needs a relatively
small number of samples.
It is able to categorize using both
binary as well as multi-systems
simultaneously.
It does not take into
consideration the
interdependencies between
characteristics for the sake of
categorization; hence, its
accuracy suffers as a result.
Accuracy=87.7%
Precision=97.7% and
F-measure=87.7%
KNN
[15]
U2R, R2L,
Flooding attacks,
DoS, DDoS
Easy to be using. The most difficult parts of this
process are finding missing
nodes as well as determining the
ideal value of K.
Recall=89.7%
Precision=90.7% and
F-measure=90.7%
DT
[16]
DDoS, U2R, R2L This method is less complicated and
easy to use.
The storage space must increase.
It is difficult to process
computationally
Recall=95.7%
Accuracy=95.7% and
F-measure=95.7%
SVM
[17]
Scan, DDoS
(TCP, UDP
flood), port sweep
SVMs may perform tasks such as
online learning and real-time
anomaly-based intruder detection.
This is made possible by their
simplicity, which makes them very
scalable.
SVMs make far less use of storage as
well as memory.
When we cannot split
information linearly, SVM is
used to classify it, but it remains
difficult to reach the needed
classification speed by
employing a perfect kernel
function.
Accuracy=88.7%
Precision=88.7% and
F-measure=88.7%
EL
[18]
DoS, Probe, R2L,
U2R attacks
It can withstand excessive or
improper fitment.
Operates at a higher level of
efficiency than a single classification.
It helps to reduce the amount of
variation.
The usage of numerous
classifiers in parallel leads to an
increase in the temporal
complexity of the process.
Recall=93.7%
Precision=93.7% and
F-measure=93.7%
RF
[19]
DoS, Probe, R2L,
U2R
It generates an output that is more
reliable and accurate, as well as one
that is sensitive to overfitting.
It takes a far lower number of inputs
than other methods, and it does away
with the need for feature selection
altogether.
Real-time programs that need a
huge dataset may not find it
feasible to employ RF since it
creates several DTs, hence this
may make its usage difficult.
Recall=92.7%
Precision=92.7% and
F-measure=92.7%
Figure 2. Intrusion detection solutions for the internet of things that make use of machine learning

1893
Figure 3. Random forest in machine learning
Figure 4. SVM in machine learning
3. APPLICATIONS OF MACHINE LEARNING IN THE INTERNET OF THINGS FOR IDS
A multi-layered deeper RNN model was presented by Almiani et al. [20] to be used for Internet of
Things devices. DoS, Probe, U2R, as well as R2L attack detection rates, were determined to be 98.26%,
97.36%, 64.94%, and 77.24% correspondingly after the performance was assessed using the NSL-KDD
database. This study presents a nearly flawless detection approach for defending against cyberattacks, as
shown by the experimental as well as performance analyses. An IDS integrated data sampling method to
solve the class imbalance is presented by Qaddoura et al. [21], this IDS proposal consists of three steps,
namely integrated cluster, classification, as well as oversampling methods. The issue of there being no
minority class is addressed via the use of oversampling. The work of Saravanan et al. [22] examines both
traditional ML methods and NIDS, as well as potential future directions. Since learning methods provide
covert security and performance in the background, we focus on IoT NIDS for Machine Learning in this
study. It offers a comprehensive look into NIDSs that use many features of internet-based learning
approaches, complementing other prominent surveys that deal with common frames. It also protects
individuals from typical IoT NIDS because of the survey's emphasis on prevention. Various kind of attacks
like phishing, probe, and eavesdropping. may take place in IoT ecosystem as shown in Figure 5.
A new concept presented by Papafotikas et al. [23], successfully detects the suspicious behavior of
smart items in an internet of things (IoT) environment, which was caused by a virus assault on an IoT node.
To spot suspicious behavior and benefit from the dissipation of current supply characteristics, machine
learning (ML) based clustering method is founded on the K-means clustering algorithm and supervised
training. The goal of this study is to develop an external current monitoring device for obtaining intrusion
detection based on the operating parameters of the equipment under monitoring. Latif et al. [24] presented a

 ISSN: 2088-8708
1894
novel technique for detecting attacks in the industrial IoT utilizing the random neural network (RNN), a
recent and lightweight ANN algorithm. Gradient descent is used to train the recommended RNN model
(GD). During dataset processing, the “Source ID” characteristic is removed to improve accuracy to greater
than 99%. Saranya et al. [25] examine the IoT, big data, smart cities, fog computing, and 5G networks as
application cases in their study of ML algorithms for intrusion detection systems (GN). To further classify
the intrusions, we employ ML approaches such as linear discriminant analysis, classification and regression
trees, and random forest. As this study shows, the detection rate, false positive rate, and accuracy all change
based on the algorithm and the application. Using a number of different types of machine learning, Shaver
et al. [26] developed an anomaly-based intrusion detection system to safeguard IoT devices. Since the
attack's signature will not be recognized, but the resulting network activity will deviate from typical patterns,
the IDS will be able to detect a zero-day attack and take corrective action. This research examines the novel
IoT network intrusion dataset. Tawalbeh et al. [27] proposed an innovative generic and expanded IoT model
with cloud/edge assistance to identify security and privacy components at each IoT layer. Amazon web
science virtual machine (AWS) was created as the IoT model's lower layer. The Raspberry Pi4 hardware
package was presented at the IoT framework's middle layer. The top layer is deployed using AWS cloud-
enabled IoT. In the described IoT cloud/edge concept, data transfer between the layers was made possible
with the use of security credentials. To solve the issues of Signature-based intrusion detection systems
(SIDS) and anomaly-based intrusion detection systems (AIDS), Khraisat et al. [28] developed a hybrid
intrusion detection system (HIDS) (anomaly-based intrusion detection). Combining a C5 classifier and a one-
class support vector classifier, HIDS provides an additional layer of protection for IoT devices. Using the
BoT-IoT dataset, the proposed HIDS is tested against a wide range of threats, including DDoS assaults, OS
and service scanning, key loggers, and data exfiltration. The problem of duplicated features has been
addressed by Albulayhi et al. [29] devised and implemented a novel approach called minimal redundancies
discriminative feature selection (MRD-FS). The distinguishing characteristics have been chosen by applying
two standards, namely, representativeness and redundancy, to the selection process. The BoT-IoT dataset
served as the basis for their model's evaluation.
𝐼𝐺 = 𝐻𝑌 − 𝐻𝑌𝑋 = 𝐻𝑋 − 𝐻(𝑋|𝑌) (1)
The expression gives the definition of 𝑦's entropy.
𝐻𝑦 = − 𝑦𝜖𝑌𝑝𝑦𝑙𝑜𝑔2(𝑝(𝑦)) (2)
Figure 5. A representation of the connections between the IoT ecosystem and possible dangers
Garg et al. [30] used techniques such as grey wolf optimizing (GWO) as well as convolutional
neural network (CNN) in the hybrid data process theory for the purpose of detecting network anomalies.
According to the authors, their model was able to obtain a higher accuracy as well as detection rate in
contrast to previous IDSs considered as the state-of-the-art. Saheed et al. [31] created an intrusion detection
system for detecting threats in IoT networks using machine learning techniques. To detect attacks in less
secure IoT networks, researchers primarily used supervised machine learning algorithms and the UNSWNB-
15 dataset. With 99.99% accuracy, the proposed model reduced communication overhead. In the future, this
model can be improved using deep learning methods on BoT-IoT datasets and can compare results to
UNSWNB-15 datasets. Table 2 shows the results and limitations of classification techniques available for
IoT threats.

1895
Table 2. Results and limitations of classification techniques available for internet of things threats
Research
Paper
Year Methodology Results Limitations
[32] 2016 The K-means clustering method
using empirical likelihood and
support vector machines
Achieving an accuracy of 96.02%,
a TP rate of 76.19%, and an F rate
of 5.92%
Having a high rate of false positives
and low true positives
[33] 2016 Connectivity systems modeled
after ant colonies that operate
autonomously
Both the DoS and Probe attacks are
highly accurate, at 99.79% and
98.55%, respectively.
The dataset used in this research is
not representative of current-day
attacks.
[34] 2017 Models using DNNs and shallow
NNs
In terms of accuracy, shallow NN
achieves 96.75%, whereas deep NN
achieves 98.27%.
As unfortunate as it is, the NSLKDD
dataset that was employed does not
adequately capture the features of
modern assaults.
[35] 2018 NB With an F-measure of 97.7 and an
accuracy of 99.3 on average, recall
is extremely high.
The created dataset's features do not
reflect real-world network behavior.
[36] 2018 ELM 83% accuracy training time is High
[37] 2019 As a means of dimensionality
reduction, we employ LDA with
NB and CF.
The false alarm rate is 5.56 percent
and the accuracy= is 84.82
High false-positive rate and low
sensitivity
[38] 2020 Decision tree The F1 score is 99.98, the accuracy
score is 99.98, the precision score is
97.38, and the recall score is 97.39.
A long period is required for model
training
4. COMPARISON OF PUBLICALLY AVAILABLE IDS DATASETS
Given the prevalence of machine learning approaches in the fight against AIDS, it is crucial to
evaluate these methods using appropriate datasets. The tabulated properties of the datasets are shown in
Table 3. According to our findings, traditional data sets like KDD'99, which were designed for wired
networks, would not help you build effective IDS for the IoT.
Saba et al. [39] proposed a two-stage hybrid model for detecting traffic offenses. At this stage, the
genetic algorithm is used to enhance the quality of the proposed model. The second step involves evaluating
the model with a battery of machine learning tools, such as support vector machines, decision trees, and
classification algorithms. Rani and Kaushal [40] proposed an efficient network intrusion detection system
based on the supervised machine learning technique of the random forest classifier and a minimum set of
features extracted from the KDDCUP99 and NSLKDD datasets. This method yields 99.5% detection
accuracy with little training time and effort required, as measured by these datasets. In order to improve upon
traditional signature-based IDS methods, an anomaly-based IDS strategy that has been evaluated using the
IoTID20 dataset [41]. This IDS method used a hybrid feature selection engine to narrow down the features to
only those that are relevant to protecting IoT devices from unauthorized access. The random was trained and
evaluated on the IoTID20 datasets. DDoS (99.95%), man-in-the-middle (MitM) (99.99%), and scanning
(99.99%) assaults are all effectively detected by this study.
Table 3. Compassion for datasets (✔ = True, ✖ = False)
Dataset Real Traffic Label data IoT traces Zero-day attacks Full packet captured Year
DARPA 98 ✔ ✔ ✖ ✖ ✔ 1998
KDDCUP 99 ✔ ✔ ✖ ✖ ✔ 1999
CAIDA ✔ ✖ ✖ ✖ ✖ 2007
NSL-KDD ✔ ✔ ✖ ✖ ✔ 2009
ISCX 2012 ✔ ✔ ✖ ✖ ✔ 2012
ADFA-WD ✔ ✔ ✖ ✔ ✔ 2017
ADFA-LD ✔ ✔ ✖ ✔ ✔ 2018
CICIDS2017 ✔ ✔ ✖ ✔ ✔ 2019
BoT-IoT ✔ ✔ ✔ ✔ ✔ 2020
5. RESULTS AND DISCUSSION
Applying the concepts of set theory (intersection and union) to the issue of feature selection
explains the result shown in Table 4. In this context, the presence of a characteristic in information gain
(IG), and gain ratio (GR) techniques (or at least one of them) is indicative of the feature's greater relevance
and non-redundancy. Multi-classification accuracy results for the IoTID20 dataset employing five distinct
ML with different feature selection (FS) techniques are given in Table 5. Our proposed feature selection
strategy improves the performance of ML classifiers, leading to the more appropriate classifications

 ISSN: 2088-8708
1896
(intersection mathematical set theory feature selection (IMF) and union mathematical set theory feature
selection (UMF)).
The results of our hybrid FS technique for multi-classification issues on the IoTID20 dataset are
displayed in Figure 6. There are five distinct ML models involved in this tactic, each of which is fused with a
unique FS method. As can be shown in Table 5 and Figure 6, our proposed model uses ensemble to detect
multi-classes with an accuracy of 99.70% utilizing either 11 or 28 features.
Table 4. Metrics for IoTID20's performance with a well curated feature set and ensemble
Feature Approach Accuracy (%) Precision (%) Recall (%) F1-Measure (%) ROC Area (%)
60 IG-Ensemble 98.80 98.8 98.8 98.8 97.5
60 GR-Ensemble 98.74 98.7 98.7 98.7 97
20 IG-Ensemble 98.73 98.7 98.7 98.7 96.9
20 GR-Ensemble 98.56 98.6 98.6 98.6 95.5
28 UMF 98.98 98.9 98.9 98.9 98.9
11 IMF 98.98 98.9 98.9 98.9 98.9
Table 5. Multi-classification accuracy on the IoTID20 dataset using the five different ML and FS approaches
FS Approach
Name
Classifiers Accuracy Classifiers Accuracy Classifiers Accuracy Classifiers Accuracy
Information Gain # Gain Ratio # Intersection # Union #
Model IG-ANN 94.6 GR-ANN 93.86 IMF-ANN 92.8 ANN UMF 94
IG-C.3.4 99.54 GR-C.3.4 99.1 IMF-C.3.4 99.59 C 4.5 UMF 99.6
IG-Bagging 99.48 GR-Bagging 98.95 IMF-Bagging 99.58 Bagging UMF 99.6
IG-kNN 98.48 GR-kNN 98.43 IMF-kNN 98.76 kNN UMK 99.1
IG-Ensemble 99.6 GR-
Ensemble
99.14 IMF-
Ensemble
99.7 Ensemble
UMF
99.7
Figure 6. Multi-classification accuracy on the IoTID20 dataset using the five different ML and FS approaches
6. CONCLUSION
As a result of the internet of things capability of transforming items from a variety of application
domains into Internet hosts, its adoption of connected devices has skyrocketed over the course of the last ten
years across all spheres of human endeavor. At approximately the same time, vulnerabilities in the IoT put
users' privacy and safety at risk. That is why it is critical to implement better IoT security measures
immediately. When it comes to securing IoT networks, a machine-learning intrusion detection solution is a
must-have. This research presents a summary of the techniques employed by IDSs to identify intrusions in
IoT networks, all of which are based on machine learning. The following part of this research is a review of
the many previous studies on this topic. Either an IDS strategy for IoT was detailed, or attack detection
techniques for IoT were offered that could be implemented into IDS. In particular, this study described the
authors' own implementations of various machine learning techniques that may be used for IDS on the
internet of things. An assessment of the data sets currently available for investigating IoT safety is also
provided. With an emphasis on intrusion detection using machine learning techniques, this study aims to
provide academics with a synthesis of the many security challenges already encountered by IoT devices and
networks.
88
90
92
94
96
98
100
102
Accuracy
ML and FS Approaches
Accuracy

1897
REFERENCES
[1] A. Hameed and A. Alomary, “Security issues in IoT: a survey,” in 2019 International Conference on Innovation and Intelligence
for Informatics, Computing, and Technologies (3ICT), Sep. 2019, pp. 1–5, doi: 10.1109/3ICT.2019.8910320.
[2] X. Ma et al., “A survey on deep learning empowered IoT applications,” IEEE Access, vol. 7, pp. 181721–181732, 2019, doi:
10.1109/ACCESS.2019.2958962.
[3] O. I. Abiodun, E. O. Abiodun, M. Alawida, R. S. Alkhawaldeh, and H. Arshad, “A review on the security of the internet of things:
challenges and solutions,” Wireless Personal Communications, vol. 119, no. 3, pp. 2603–2637, Aug. 2021, doi: 10.1007/s11277-
021-08348-9.
[4] T. U. Sheikh, H. Rahman, H. S. Al-Qahtani, T. K. Hazra, and N. U. Sheikh, “Countermeasure of attack vectors using signature-
based IDS in IoT environments,” in 2019 IEEE 10th Annual Information Technology, Electronics and Mobile Communication
Conference (IEMCON), Oct. 2019, pp. 1130–1136, doi: 10.1109/IEMCON.2019.8936231.
[5] M. M. Noor and W. H. Hassan, “Current research on internet of things (IoT) security: a survey,” Computer Networks, vol. 148,
pp. 283–294, Jan. 2019, doi: 10.1016/j.comnet.2018.11.025.
[6] G. R. Kumar, N. Mangathayaru, and G. Narsimha, “Intrusion detection a text mining based approach,” Arxiv.org/abs/1603.03837,
Mar. 2016.
[7] S. Smys, A. Basar, and H. Wang, “Hybrid intrusion detection system for internet of things (IoT),” Journal of ISMAC, vol. 2,
no. 4, pp. 190–199, Sep. 2020, doi: 10.36548/jismac.2020.4.002.
[8] S. Deshmukh-Bhosale and S. S. Sonavane, “A real-time intrusion detection system for wormhole attack in the RPL based internet
of things,” Procedia Manufacturing, vol. 32, pp. 840–847, 2019, doi: 10.1016/j.promfg.2019.02.292.
[9] S. Rathore and J. H. Park, “Semi-supervised learning based distributed attack detection framework for IoT,” Applied Soft
Computing, vol. 72, pp. 79–89, Nov. 2018, doi: 10.1016/j.asoc.2018.05.049.
[10] C. Lyu, X. Zhang, Z. Liu, and C.-H. Chi, “Selective authentication based geographic opportunistic routing in wireless sensor
networks for internet of things against DoS attack,” IEEE Access, vol. 7, pp. 31068–31082, 2019, doi:
10.1109/ACCESS.2019.2902843.
[11] N. Moustafa, K. K. R. Choo, I. Radwan, and S. Camtepe, “Outlier Dirichlet mixture mechanism: adversarial statistical learning
for anomaly detection in the fog,” IEEE Transactions on Information Forensics and Security, vol. 14, no. 8, pp. 1975–1987, 2019,
doi: 10.1109/TIFS.2018.2890808.
[12] A. Sagu and N. S. Gill, “Machine learning techniques for securing IoT environment,” International Journal of Innovative
Technology and Exploring Engineering, vol. 9, no. 4, pp. 977–982, Feb. 2020, doi: 10.35940/ijitee.D1209.029420.
[13] Y. Sei, J. A. Onesimu, and A. Ohsuga, “Machine learning model generation with copula-based synthetic dataset for local
differentially private numerical data,” IEEE Access, vol. 10, pp. 101656–101671, 2022, doi: 10.1109/ACCESS.2022.3208715.
[14] S. Garg, K. Kaur, G. Kaddoum, F. Gagnon, N. Kumar, and Z. Han, “Sec-IoV: a multi-stage anomaly detection scheme for internet
of vehicle,” in Proceedings of the ACM MobiHoc Workshop on Pervasive Systems in the IoT Era, Jul. 2019, pp. 37–42, doi:
10.1145/3331052.3332476.
[15] J. M. Torres, C. Iglesias Comesaña, and P. J. García-Nieto, “Review: machine learning techniques applied to cybersecurity,”
International Journal of Machine Learning and Cybernetics, vol. 10, no. 10, pp. 2823–2836, Oct. 2019, doi: 10.1007/s13042-018-
00906-1.
[16] C. Ioannou and V. Vassiliou, “Classifying security attacks in IoT networks using supervised learning,” in 2019 15th International
Conference on Distributed Computing in Sensor Systems (DCOSS), May 2019, pp. 652–658, doi: 10.1109/DCOSS.2019.00118.
[17] P. Illy, G. Kaddoum, C. Miranda Moreira, K. Kaur, and S. Garg, “Securing fog-to-things environment using intrusion detection
system based on ensemble learning,” in 2019 IEEE Wireless Communications and Networking Conference (WCNC), Apr. 2019,
pp. 1–7, doi: 10.1109/WCNC.2019.8885534.
[18] D. H. Hoang and H. D. Nguyen, “Detecting anomalous network traffic in IoT networks,” in 2019 21st International Conference
on Advanced Communication Technology (ICACT), Feb. 2019, pp. 1143–1152, doi: 10.23919/ICACT.2019.8702032.
[19] P. Maniriho, E. Niyigaba, Z. Bizimana, V. Twiringiyimana, L. J. Mahoro, and T. Ahmad, “Anomaly-based intrusion detection
approach for IoT networks using machine learning,” in 2020 International Conference on Computer Engineering, Network, and
Intelligent Multimedia (CENIM), Nov. 2020, pp. 303–308, doi: 10.1109/CENIM51130.2020.9297958.
[20] M. Almiani, A. AbuGhazleh, A. Al-Rahayfeh, S. Atiewi, and A. Razaque, “Deep recurrent neural network for IoT intrusion
detection system,” Simulation Modelling Practice and Theory, vol. 101, May 2020, doi: 10.1016/j.simpat.2019.102031.
[21] R. Qaddoura, A. M. Al-Zoubi, I. Almomani, and H. Faris, “A multi-stage classification approach for IoT intrusion detection based
on clustering with oversampling,” Applied Sciences, vol. 11, no. 7, Mar. 2021, doi: 10.3390/app11073022.
[22] M. Saravanan, A. M. Thoufeeq, S. Akshaya, and V. L. J. Manchari, “Exploring new privacy approaches in a scalable
classification framework,” in 2014 International Conference on Data Science and Advanced Analytics (DSAA), Oct. 2014,
pp. 209–215, doi: 10.1109/DSAA.2014.7058075.
[23] S. Papafotikas and A. Kakarountas, “A machine-learning clustering approach for intrusion detection to IoT devices,” in 2019 4th
South-East Europe Design Automation, Computer Engineering, Computer Networks and Social Media Conference (SEEDA-
CECNSM), Sep. 2019, pp. 1–6, doi: 10.1109/SEEDA-CECNSM.2019.8908520.
[24] S. Latif, Z. Zou, Z. Idrees, and J. Ahmad, “A novel attack detection scheme for the industrial internet of things using a lightweight
random neural network,” IEEE Access, vol. 8, pp. 89337–89350, 2020, doi: 10.1109/ACCESS.2020.2994079.
[25] T. Saranya, S. Sridevi, C. Deisy, T. D. Chung, and M. K. A. A. Khan, “Performance analysis of machine learning algorithms in
intrusion detection system: a review,” Procedia Computer Science, vol. 171, pp. 1251–1260, 2020, doi: 10.1016/j.procs.2020.04.133.
[26] A. Shaver, Z. Liu, N. Thapa, K. Roy, B. Gokaraju, and X. Yuan, “Anomaly based intrusion detection for iot with machine
learning,” in 2020 IEEE Applied Imagery Pattern Recognition Workshop (AIPR), Oct. 2020, pp. 1–6, doi:
10.1109/AIPR50011.2020.9425199.
[27] L. Tawalbeh, F. Muheidat, M. Tawalbeh, and M. Quwaider, “IoT privacy and security: challenges and solutions,” Applied
Sciences, vol. 10, no. 12, Jun. 2020, doi: 10.3390/app10124102.
[28] A. Khraisat, I. Gondal, P. Vamplew, J. Kamruzzaman, and A. Alazab, “A novel ensemble of hybrid intrusion detection system for
detecting internet of things attacks,” Electronics, vol. 8, no. 11, Oct. 2019, doi: 10.3390/electronics8111210.
[29] K. Albulayhi and F. T. Sheldon, “An adaptive deep-ensemble anomaly-based intrusion detection system for the internet of
things,” in 2021 IEEE World AI IoT Congress (AIIoT), May 2021, pp. 187–196, doi: 10.1109/AIIoT52608.2021.9454168.
[30] S. Garg, K. Kaur, N. Kumar, G. Kaddoum, A. Y. Zomaya, and R. Ranjan, “A hybrid deep learning-based model for anomaly
detection in cloud datacenter networks,” IEEE Transactions on Network and Service Management, vol. 16, no. 3, pp. 924–935,
Sep. 2019, doi: 10.1109/TNSM.2019.2927886.
[31] Y. K. Saheed, A. I. Abiodun, S. Misra, M. K. Holone, and R. Colomo-Palacios, “A machine learning-based intrusion detection for

 ISSN: 2088-8708
1898
detecting internet of things network attacks,” Alexandria Engineering Journal, vol. 61, no. 12, pp. 9395–9409, Dec. 2022, doi:
10.1016/j.aej.2022.02.063.
[32] H. Bostani and M. Sheikhan, “Hybrid of anomaly-based and specification-based IDS for Internet of Things using unsupervised
OPF based on MapReduce approach,” Computer Communications, vol. 98, pp. 52–71, Jan. 2017, doi:
10.1016/j.comcom.2016.12.001.
[33] Y. Feng, J. Zhong, C. Ye, and Z. Wu, “Clustering based on self-organizing ant colony networks with application to intrusion
detectio,” in Sixth International Conference on Intelligent Systems Design and Applications, Oct. 2006, vol. 2, pp. 1077–1080,
doi: 10.1109/ISDA.2006.253761.
[34] A. A. Diro and N. Chilamkurti, “Distributed attack detection scheme using deep learning approach for internet of things,” Future
Generation Computer Systems, vol. 82, pp. 761–768, May 2018, doi: 10.1016/j.future.2017.08.043.
[35] E. Anthi, L. Williams, and P. Burnap, “Pulse: an adaptive intrusion detection for the internet of things,” 2018, doi:
10.1049/cp.2018.0035.
[36] R. Kozik, M. Choraś, M. Ficco, and F. Palmieri, “A scalable distributed machine learning approach for attack detection in edge
computing environments,” Journal of Parallel and Distributed Computing, vol. 119, pp. 18–26, Sep. 2018, doi:
10.1016/j.jpdc.2018.03.006.
[37] H. H. Pajouh, R. Javidan, R. Khayami, A. Dehghantanha, and K. K. R. Choo, “A two-layer dimension reduction and two-tier
classification model for anomaly-based intrusion detection in IoT backbone networks,” IEEE Transactions on Emerging Topics in
Computing, vol. 7, no. 2, pp. 314–323, 2019, doi: 10.1109/TETC.2016.2633228.
[38] Y.-W. Chen, J.-P. Sheu, Y.-C. Kuo, and N. Van Cuong, “Design and implementation of IoT DDoS attacks detection system based
on machine learning,” in 2020 European Conference on Networks and Communications (EuCNC), Jun. 2020, pp. 122–127, doi:
10.1109/EuCNC48522.2020.9200909.
[39] T. Saba, T. Sadad, A. Rehman, Z. Mehmood, and Q. Javaid, “Intrusion detection system through advance machine learning for
the internet of things networks,” IT Professional, vol. 23, no. 2, pp. 58–64, Mar. 2021, doi: 10.1109/MITP.2020.2992710.
[40] D. Rani and N. C. Kaushal, “Supervised machine learning based network intrusion detection system for internet of things,” in
2020 11th International Conference on Computing, Communication and Networking Technologies (ICCCNT), Jul. 2020, pp. 1–7,
doi: 10.1109/ICCCNT49239.2020.9225340.
[41] B. Susilo and R. F. Sari, “Intrusion detection in IoT networks using deep learning algorithm,” Information, vol. 11, no. 5, May
2020, doi: 10.3390/info11050279.
BIOGRAPHIES OF AUTHORS
Nisha received her BCA degree in computer science from Maharshi Dayanand
University, Rohtak (Haryana), and received her MCA degree from Guru Gobind Singh
Indraprastha University Delhi. She is a Ph.D. scholar at the Department of Computer Science
and Applications, Maharshi Dayanand University, Rohtak, Haryana, India. She has qualified
National Eligibility Test for assistant professor in India. She may be contacted at email:
nisha.rs.dcsa@mdurohtak.ac.in.
Nasib Singh Gill is currently head of Department of Computer Science and
Applications, M.D. University, Rohtak, India. He is also working as Director, Directorate of
Distance Education as well as director of digital learning center, M.D. University, Rohtak,
Haryana. He earned his doctorate in computer science in the year 1996 and carried out his
post-doctoral research at Brunel University, West London during 2001-2002. He is a
recipient of the Commonwealth Fellowship Award of the British Government for the Year
2001. Besides, he also has earned his MBA degree. He is an active professional member of
IETE, IAENG, and CSI. He has published more than 304 research papers and authored 5
popular books. He has guided so far 12 Ph.D. scholars as well as guiding about 5 more
scholars. His research interests primarily include IoT, machine and deep learning,
information and network security, data mining and data warehousing, NLP, and measurement
of component-based systems. He can be contacted at email: nasib.gill@mdurohtak.ac.in.
Preeti Gulia is currently working as an associate professor at the Department of
Computer Science and Applications, M.D. University, Rohtak, India. She has been serving
the Department since 2009. She earned her doctoral degree in 2013. She has published more
than 65 research papers and articles in journals and conferences of National/International
repute including ACM, and Scopus. Her area of research includes data mining, big data,
machine learning, deep learning, IoT, and software engineering. She is an active professional
member of IAENG, CSI, and ACM. She is also serving as an editorial board member active
reviewer of international/national journals. She has guided four research scholars as well as
guiding six Ph.D. research scholars from various research areas at present. She can be
contacted at email: preeti@mdurohtak.ac.in.

A review on machine learning based intrusion detection system for internet of things enabled environment

More Related Content

Similar to A review on machine learning based intrusion detection system for internet of things enabled environment (20)

More from IJECEIAES (20)

Recently uploaded (20)

A review on machine learning based intrusion detection system for internet of things enabled environment