Aparcana-TasaycoAndres J.1
Gamboa-CruzadoJavier1
-
(Escuela Profesional de Ingenieria de Sistemas, Universidad Autonoma del Peru / Lima,
Peru {aaparcana, jgamboa}@autonoma.edu.pe
)
Copyright © The Institute of Electronics and Information Engineers(IEIE)
Keywords
Machine learning (ML), Management, Software-defined networks (SDN), Systematic literature review (SLR)
1. Introduction
The Software-Defined Networking (SDN) paradigm has emerged as a response to the inability
of today's networks to maintain a global view of network elements [87]. SDN enables innovation and programmability through a centralized structure, which
in turn provides global-view data that can be used by different algorithms [89]. Many technologies have been used to improve various aspects of management in software-defined
networks, and one of the most important is Machine Learning (ML) [89]. With these techniques, networks become smarter, more autonomous, and more efficient
in improving the utilization and optimization of network resources [88].
Research linking two variables has increased in the last five years. This makes it
necessary to have a systematic literature review that brings together this research
and answers the right questions for researchers in the area, as well as comparing
it with similar works. Thus, a systematic literature review was done to analyze ML
and its influence on SDN management.
The paper is structured as follows. Section 2 presents a summary of the concepts used
and related works that serve as a comparison with the present study. Section 3 explains
the methodology used for the systematic literature review based on Kitchenham’s guidelines
[82] and the PRISMA guidelines. Section 4 analyzes and compares the results obtained with
those of other researchers. Finally, Section 5 presents the conclusions of the study
and guidelines for future research.
2. Background and Related Works
In the process of systematic literature review, the search for review papers was replicated
to compare the results obtained. However, in many of the cases, the papers did not
make very broad explorations of areas within the software-defined networks, so the
comparisons could not be made completely. The studies were focused on certain areas
applicable to software-defined networks. However, they were not limited to studies
in this paradigm of networks. Also, these studies present descriptive questions that
are much simpler than the analytics presented in this paper. It is important to highlight
that these investigations were from 2019 onward, so the growing demand for these papers
still leaves room for future research.
In 2019, Öney and Peker [86] published a review on only techniques based on neural networks and their influence
on Intrusion Detection Systems (IDS). The results of this research were mainly that
the most used technique to deal with IDS was deep neural networks, and the main database
was KDD'99. Then, in 2020, Amarudin, Ferdiana, and Widyawan [85] published a study that also focused on IDSs for network security and analyzed the
papers related to techniques in ML. As a result, the SVM technique was the most used
to give IDS solutions, and 87 % of the ML techniques studied were classification techniques.
In the year 2021, there were two important studies prepared by Nassif et al. [83,84], who deeply studied the techniques of ML and the application for anomaly detection
and security in the Cloud, respectively. The first paper [84] highlights increasing research in the area in recent years. The date range used by
the authors was 20 years. It also highlighted that the SVM technique is the most used
for anomaly detection but that unsupervised techniques are the most abundant among
researchers.
The second study [83] extended the studies to cloud security. They highlight that in this field, DDoS attacks
are the most common and also reaffirm that SVM is the most used technique in the literature.
They particularly mention that KDD and KDD CUP'99 are the most commonly used databases
to enter algorithms. They recommend that researchers include data selection and extraction
strategies in the systematic reviews consulted. Finally, they also recommend conducting
research in this area.
The contribution of this work is to expand knowledge and analyze ML techniques for
DN Management. In the present study, the following contributions are made. The search
equations follow strict PRISMA guidelines to answer the research questions, the search
sources are specialized and include peer-reviewed articles, and the author makes use
of the Research Assistant Joshua (RAj) tool, an intelligent assistant developed by
Gamboa-Cruzado. RAj makes use of the TextBlob release for sentiment analysis. Polarity
refers to the feeling of the text, where -1.0 is a text written in a very negative
way, 0.0 is neutral, and 1.0 is very positive. Objectivity refers to how the text
is written, where 0.0 is given to a text written entirely based on facts, and 1.0
is given to a text written entirely based on opinions and personal feelings [90].
3. Methodology
The methodology of the systematic review of literature takes into account the work
of Kitchenham [82]. The methodology defines research phases and techniques involving research problems
and objectives, search sources and search strategies, identified studies, exclusion
criteria, paper selection, quality assessment, data extraction strategy, and finally,
data synthesis.
3.1 Research Problems and Objectives
Identifying research questions and objectives is an essential procedure in the systematic
literature review process because it is necessary for the following steps: search
strategy planning, data extraction, and data analysis. Table 1 shows a table with the research questions and their related objectives.
Table 1. Research questions and objectives.
|
Research Question
|
Objectives
|
|
RQ1: Who are the most productive authors in ML development?
|
Identify the most productive authors in ML development
|
|
RQ2: What are the most cited articles on ML and its influence on software-defined
network management whose discussions and conclusions are characterized by their objectivity
and polarity?
|
Identify the most cited articles on ML and its influence on software-defined network
management whose discussions and conclusions are characterized by their objectivity
and polarity
|
|
RQ3: Who are the authors who are frequently Co-authors in research on ML and its influence
on software-defined network management?
|
Determine the authors who are often co-authors in research on ML and its influence
on software-defined network management
|
|
RQ4: What are the keywords that frequently present cooccurrence in ML research and
its influence on software-defined network management?
|
Identify the Keywords that frequently present cooccurrence in ML research and its
influence on software-defined network management
|
3.2 Search Sources and Search Strategies
Scientific data sources were chosen to provide research papers focused on engineering,
which were used to find the papers relevant to the study. The sources were Taylor
& Francis, IEEE Xplore, Springer Link, Scopus, Science Direct, Wiley Online Library,
ACM Digital Library, and ProQuest. The search strategy requires the identification
of descriptors related to the independent and dependent variables. Table 2 shows the descriptors, which make use of synonyms separated by "/" to expand and
consider similar papers that do not have nomenclatures equal to those of the study.
The search procedure was carried out using search equations that strictly respect
the syntax of the chosen data sources. These are shown in Table 3.
Table 2. Search descriptors.
|
Descriptor
|
Variable
|
|
Machine learning / ML
|
Independent
|
|
Network management / management
|
Dependent
|
|
Software-defined networks / SDN
|
|
Methodology / Method / Model
|
Table 3. Sources and search equations.
|
Sources
|
Search equation
|
|
Taylor & Francis
|
[[All: "machine learning"] OR [All: ml]] AND [[All: management] OR [All: "network
management"]] AND [[All: "software-defined network"] OR [All: sdn]] AND [[All: methodology]
OR [All: method] OR [All: model]]
|
|
IEEE Xplore
|
(("Full Text & Metadata":"Machine Learning") OR ("Full Text & Metadata":ML)) AND (("Full
Text & Metadata":Management) OR ("Full Text & Metadata":"Network Management")) AND
(("Full Text & Metadata":"Software-Defined Network") OR ("Full Text & Metadata":SDN))
AND (("Full Text & Metadata":Methodology) OR ("Full Text & Metadata":Method) OR ("Full
Text & Metadata":Model))
|
|
Springer Link
|
("Machine Learning" OR ML) AND (Management OR "Network Management") AND ("Software-Defined
Network" OR SDN) AND (Methodology OR Method OR Model)
|
|
Scopus
|
ALL ( ( "Machine Learning" OR ML ) AND ( Management OR "Network Management"
) AND ( "Software-Defined Network" OR SDN ) AND ( Methodology OR Method OR
Model ) )
|
|
Science Direct
|
("Machine Learning" OR ML) AND (Management OR "Network Management") AND ("Software-Defined
Network" OR SDN) AND (Methodology OR Method OR Model)
|
|
Wiley Online Library
|
""Machine Learning" OR ML" anywhere and "Management OR "Network Management"" anywhere
and ""Software-Defined Network" OR SDN" anywhere and "Methodology OR Method OR Model"
anywhere
|
|
ACM Digital Library
|
[[All: "machine learning"] OR [All: ml]] AND [[All: management] OR [All: "network
management"]] AND [[All: "software-defined network"] OR [All: sdn]] AND [[All: methodology]
OR [All: method] OR [All: model]]
|
|
ProQuest
|
("Machine Learning" OR ML) AND ( Management OR "Network Management" ) AND
( "Software-Defined Network" OR SDN ) AND ( Methodology OR Method OR Model)
|
3.3 Identified Studies
Once the search for research papers is complete, the results are presented in a diagram
that includes the results from data sources and the total set of primary papers found.
This is shown in Fig. 1.
3.4 Exclusion Criteria
Exclusion criteria were defined to filter the papers relevant to the research. Given
the criteria below, the papers were filtered in the following order:
EC1. The items are older than 5 years.
EC2. Papers are not written in English.
EC3. Papers were not published in peer-reviewed conferences or journals.
EC4. Paper titles and keywords are not very suitable.
EC5. The proposed solution does not apply to software-defined network management.
Fig. 1. Number of studies identified.
EC6. The abstract of the papers is not very relevant.
EC7. The items are not unique.
EC8. The full text of the paper is not available.
3.5 Selection of Papers
The first results gave 21,743 papers based on the strategy described in section 3.2
using the descriptors of the study. The following selection and filtering steps were:
1) Applying exclusion criteria to guarantee relevant documents for the study
2) Applying quality assessments to include papers that give a clear answer to the
given research questions
The results of the first step gave the 81 papers shown in Fig. 2 (PRISMA chart).
Fig. 2. Application of exclusion criteria.
3.6 Quality Assessment
Applying quality assessment was the second step described in section 3.5 to identify
a final set of papers that were included in this systematic review. QAs have been
applied to evaluate the quality of the papers. 7 QAs were selected, which are listed
as follows:
QA1. Are the research objectives visibly identified in the paper?
QA2. Is the experiment appropriate and acceptable?
QA3. Does the paper explain the context in which the research was conducted?
QA4. Is the document well constituted?
QA5. Are the methods used to analyze the results timely?
QA6. Are the results of the experiment visibly identified and reported?
QA7. Is the document considered appropriate?
After an exhaustive review of the filtered papers, we verified the quality based on
the criteria presented above to ensure their relevance within the objectives set out
in section 3. It was concluded that all 81 studies met the quality criteria.
3.7 Data Extraction Strategy
A data collection structure was used where the necessary information is extracted
to answer the research questions. The information extracted from each paper includes
the following data: PAPER ID, title, URL, source, year, country, number of pages,
language, type of paper, publication medium name, authors, affiliation, number of
citations, abstract, keywords, sample size, RQ1, RQ2, RQ3, and RQ4. The Zotero software
was used to obtain the data, as shown in Fig. 3.
Fig. 3. Document management with Zotero.
3.8 Data Synthesis
Finally, the data were collected, typed, and presented in figures and tables to make
a visual analysis and a statistical comparison of the results obtained. These figures
and tables answer each research question posed and show relevant findings from papers
published during the last five years.
4. Results and Discussion
4.1 Study Overview
The selection ended with 81 research papers that were rigorously analyzed to obtain
the relevant data for the study. The papers were selected considering the last five
years; that is, papers from 2016 to the year of 2021. Fig. 4 shows the distribution of papers in these years, and the tendency to use ML techniques
in the management of software-defined networks is growing. It is unavoidable that
ML techniques are becoming increasingly important in research and development within
the paradigm of software-defined networks.
Fig. 5 shows the location of the most relevant papers for the study. This figure allows
researchers to define the countries that invest the most in research in software-defined
networks by applying ML. The countries with the most research are China, the United
States, and India with 15 (18.52 %), 14 (17.28 %), and 11 (13.58 %) publications,
respectively. The selected papers also showed remarkable results in terms of the ML
algorithms (Table 4) used by the researchers. Among these algorithms, Support Vector Machine (SVM), Naive
Bayes, and Reinforcement Learning are the most popular with 14.4 %, 9.3 %, and 7.9
%, respectively.
Although it is not included as a research question, it is important to know what techniques
are used in the reviewed papers. The review shows us that SVM is the most popular
ML algorithm. This is partially confirmed in
Fig. 5. Georeferential map of publications by country.
the reviews consulted since they do not focus on SDN only. According to Nassif, Talib,
Nasir, and Dakalbab [84], for Anomaly Detection, SVM is the most popular ML technique. In another study on
Cloud Security, the same authors [83] reveal that SVM is still the first technique. Amarudin, Ferdiana, and Widyawan [85] focused on IDSs and stated that SVM maintains the first place with the same number
of studies. Finally, Öney and Peker [86] focused on IDSs with only Neural Network (NN) techniques contradict the results obtained,
finding deep neural networks to have the largest number of studies.
4.2 Answers to Research Questions
RQ1: Who are the most productive authors in ML development?
Given the question asked, the study has done the task of finding the most productive
authors in ML. Fig. 6 shows that Julong Lan and Yuxiang Hu are the most productive authors, both with 5
papers. Secondly, Penghao Sun and Zehua Guo have 4 items. It should be noted that
the first two authors have the same papers, which were mostly indexed in Scopus. Likewise,
the second place also has papers found in this data source, so these authors publish
high-quality papers.
Undoubtedly, the most productive authors publish in data sources indexed in Scopus.
The relationship between these authors is discussed in question RQ4. This answer fulfills
the objective of identifying the authors with more products in the area but does not
find review papers that have asked the same question.
RQ2: What are the most cited papers on ML and its influence on software-defined network
management whose discussions and conclusions are characterized by their objectivity
and polarity?
Based on this question, Fig. 7 shows the objectivity and polarity of the discussions and conclusions ordered by
the most cited papers. It can be noted that the most cited paper is subjective and
of positive polarity. Positive polarity means that the paper was written formally.
Instead, the following papers are shown to be neutral in both their objectivity and
polarity. The most cited papers are from the scientific data sources IEEE Xplore,
Scopus, and ACM Digital Library. All papers are also shown in a graph of objectivity
and polarity in Fig. 8.
The study finds that for this question most of the most cited papers are in the balance
of these characteristics. These papers maintain a factual argument of their findings
in their research, and they also write neutrally to the reader. The study determined
the most cited papers and characterized them for their objectivity and polarity. No
other review paper consulted contains any similar questions for comparison.
RQ3: Who are the authors who are frequently co-authors in research on ML and its influence
on software-defined network management?
Authors who are often co-authors in research on ML and software-defined network management
are shown visually with bibliometric networks in Fig. 9. The elaboration of these bibliometric networks was done while taking into account
that authors and their co-authors carry out more research. Consequently, the authors
who appear are also the most productive, as answered in RQ1. The figure mainly shows
that Julong Lan and Yuxian Hu share the same publications (5). It is also appreciated
that these authors carry out work together with other researchers, such as Penghao
Sun sharing 7 papers through a network of researchers that link Junfei Li and Zehua
Guo. They make up the strongest network of authors.
The present study presents graphs not found in the systematic reviews consulted. The
graphs are bibliometric networks that visually show the strongest relationships between
the authors, who are also the ones with the most publications. These authors are of
Chinese origin and belong to the same research institutes.
RQ4: What are the Keywords that frequently present cooccurrence in ML research and
its influence on software-defined network management?
To answer this question, two figures were elaborated. Fig. 10 shows that the most frequent keywords are "Machine Learning", "SDN", and as a complete
concept, "Software-Defined Networking". This is mainly due to the rigorous selection
of research variables. These more frequent keywords also show important connections.
Fig. 11 shows bibliometric networks that relate the most important keywords. These networks
show that "Machine Learning" has a strong relationship with "SDN" and "Software-Defined
Networking". It also shows strong relationships with the most used ML techniques,
as well as important concepts in SDN, such as the OpenFlow protocol and security.
The last question has also been answered by bibliometric networks that identify the
strong relationships between the words "Machine Learning" and "Software-Defined Networks",
or its more popular abbreviation, "SDN". These types of questions and graphical representations
are also not found in the systematic reviews consulted.
Fig. 6. Most productive authors in ML.
Fig. 7. Most cited papers characterized by their objectivity and polarity.
Fig. 8. Papers in graph of objectivity and polarity.
Fig. 9. Bibliometric networks of authors.
Fig. 10. Most used keywords.
Fig. 11. Bibliometric network of keywords.
5. Conclusions and Future Research
The relationship between ML and its influence on software-defined network management
has increased in the number of publications in the last five years. In this study,
81 papers were determined to contribute to the 4 research questions posed. These papers
were chosen after a search in the most specialized scientific data sources on these
topics: Taylor & Francis, IEEE Xplore, Springer Link, Scopus, Science Direct, Wiley
Online Library, ACM Digital Library, and ProQuest. Exclusion and quality criteria
were also applied to them through the adoption of Kitchenham’s guidelines [82], and together with the Zotero tool, the correct data can be extracted.
The results and discussion showed important findings. The most productive countries
are China, US, and India, and the most popular technique among researchers is SVM.
In response to RQ1, Julong Lan and Yuxian Hu are the authors with the most publications.
The response to RQ2 shows that the most cited papers are mostly neutral. For RQ3,
the result relates authors and co-authors, which gives us a look at the academic relationships
of authors and products. Finally, the answer of RQ4 gives bibliometric networks of
the most used keywords and those with whom they relate the most.
Researchers in ML and its influence on software-defined network management are recommended
to consider the evolution of concepts in variables their review papers, as well as
key phrases that influence this type of research. Upcoming researchers are also urged
to use bibliometric networks in research questions that analyze the relationships
between elements through the use of IT tools. Finally, it is proposed that in future
systematic reviews, ML and text mining techniques be implemented to increase the objectivity
and polarity of the investigations.
ACKNOWLEDGMENTS
This research was supported by Universidad Autonoma del Peru, Lima, Peru. We thank
Prof. Favio Cuya for his invaluable help during the review process.
REFERENCES
Alshamrani A., Chowdhary A., Pisharody S., Lu D., Huang D., 2017, A Defense System
for Defeating DDoS Attacks in SDN based Networks, Proceedings of the 15th ACM International
Symposium on Mobility Management and Wireless Access. Association for Computing Machinery,
Miami, Florida, USA, pp. 83-92
Prakash A., Priyadarshini R., Apr. 2018, An Intelligent Software defined Network Controller
for preventing Distributed Denial of Service Attack, in 2018 Second International
Conference on Inventive Communication and Computational Technologies (ICICCT), Coimbatore,
pp. 585-589
Lazaris A., Prasanna V. K., Mar. 2021, An LSTM Framework for Software-Defined Measurement,
IEEE Trans. Netw. Serv. Manage., Vol. 18, No. 1, pp. 855-869
Lazaris A., Prasanna V. K., 2017, DeepFlow: a deep learning framework for software-defined
measurement, in Proceedings of the 2nd Workshop on Cloud-Assisted Networking - CAN
'17, Incheon, Republic of Korea, pp. 43-48
El-Shamy A. M., El-Fishawy N. A., Attiya G., Mohamed M. A. A., Jan. 2021, Anomaly
Detection and Bottleneck Identification of The Distributed Application in Cloud Data
Center using Software-Defined Networking, Egyptian Informatics Journal
Rego A., Sendra S., Garcia L., Lloret J., Sep. 2019, Adapting reinforcement learning
for multimedia transmission on SDN, Trans Emerging Tel Tech, Vol. 30, No. 9
Canovas A., Rego A., Romero O., Lloret J., Jan. 2020, A robust multimedia traffic
SDN-Based management system using patterns and models of QoE estimation with BRNN,
Journal of Network and Computer Applications, Vol. 150, pp. 102498
Ruelas A. M. R., Rothenberg C. E., Oct. 2018, A Load Balancing Method based on Artificial
Neural Networks for Knowledge-defined Data Center Networking, in Proceedings of the
10th Latin America Networking Conference, São Paulo Brazil, pp. 106-109
Pradana A. M., Purboyo T. W., Latuconsina R., 2020, A Simulation Of Load Balancing
In Software Defined Network (Sdn) Based On Artificial Neural Networks Method, ARPN
Journal of Engineering and Applied Sciences, Vol. 15, No. 6, pp. 748-758
Volkov A., Proshutinskiy K., Adam A. B. M., Ateya A. A., Muthanna A., Koucheryavy
A., 2019, SDN Load Prediction Algorithm Based on Artificial Intelligence, in Distributed
Computer and Communication Networks, Cham, pp. 27-40
Ben Letaifa A., Maher G., Mouna S., Jun. 2017, ML based QoE enhancement in SDN context:
Video streaming case, in 2017 13th International Wireless Communications and Mobile
Computing Conference (IWCMC), Valencia, Spain, pp. 103-108
Abubakar A., Pranggono B., Sep. 2017, Machine learning based intrusion detection system
for software defined networks, in 2017 Seventh International Conference on Emerging
Security Technologies (EST), Canterbury, pp. 138-143
Kyaw A. T., Zin Oo M., Khin C. S., Jun. 2020, Machine-Learning Based DDOS Attack Classifier
in Software Defined Network, in 2020 17th International Conference on Electrical Engineering/Electronics,
Computer, Telecommunications and Information Technology (ECTI-CON), Phuket, Thailand,
pp. 431-434
Yu C., Lan J., Guo Z., Hu Y., 2018, DROM: Optimizing the Routing in Software-Defined
Networks With Deep Reinforcement Learning, IEEE Access, Vol. 6, pp. 64533-64539
Liu C.-C., Chang Y., Tseng C.-W., Yang Y.-T., Lai M.-S., Chou L.-D., Jul. 2018, SVM-based
Classification Mechanism and Its Application in SDN Networks, in 2018 10th International
Conference on Communication Software and Networks (ICCSN), Chengdu, pp. 45-49
Zhang C., Wang X., Li F., He Q., Huang M., May 2018, Deep learning-based network application
classification for SDN: Deep learning-based network application classification for
SDN, Trans Emerging Tel Tech, Vol. 29, No. 5, pp. e3302
Zerbini C. B., Carvalho L. F., Abrão T., Proença M. L., Jul. 2019, Wavelet against
random forest for anomaly mitigation in software-defined networking, Applied Soft
Computing, Vol. 80, pp. 138-153
Khamaiseh S., Serra E., Li Z., Xu D., Oct. 2019, Detecting Saturation Attacks in SDN
via Machine Learning, in 2019 4th International Conference on Computing, Communications
and Security (ICCCS), Rome, Italy, pp. 1-8
Castillo E. F., Rendon O. M. C., Ordonez A., Zambenedetti Granville L., Apr. 2020,
IPro: An approach for intelligent SDN monitoring, Computer Networks, Vol. 170, pp.
107108
S. E. R., Ravi R., Mar. 2020, A performance analysis of Software Defined Network based
prevention on phishing attack in cyberspace using a deep machine learning with CANTINA
approach (DMLCA), Computer Communications, Vol. 153, pp. 375-381
Bouzidi E. H., Outtagarts A., Langar R., Dec. 2019, Deep Reinforcement Learning Application
for Network Latency Management in Software Defined Networks, in 2019 IEEE Global Communications
Conference (GLOBECOM), Waikoloa, HI, USA, pp. 1-6
Unal E., Sen-Baidya S., Hewett R., Dec. 2018, Towards Prediction of Security Attacks
on Software Defined Networks: A Big Data Analytic Approach, in 2018 IEEE International
Conference on Big Data (Big Data), pp. 4582-4588
Alhaidari F., et al. , 2021, Intelligent Software-Defined Network for Cognitive Routing
Optimization using Deep Extreme Learning Machine Approach, Computers, Materials &
Continua, Vol. 67, No. 1, pp. 1269-1285
Benayas F., Carrera A., Iglesias C. A., Apr. 2018, Towards an autonomic Bayesian fault
diagnosis service for SDN environments based on a big data infrastructure, in 2018
Fifth International Conference on Software Defined Systems (SDS), pp. 7-13
Kaur G., Gupta P., Aug. 2019, Hybrid Approach for detecting DDOS Attacks in Software
Defined Networks, in 2019 Twelfth International Conference on Contemporary Computing
(IC3), Noida, India, pp. 1-6
Cusack G., Michel O., Keller E., 2018, Machine Learning-Based Detection of Ransomware
Using SDN, Proceedings of the 2018 ACM International Workshop on Security in Software
Defined Networks & Network Function Virtualization. Association for Computing Machinery,
Tempe, AZ, USA, pp. 1-6
Alamri H. A., Thayananthan V., 2020, Bandwidth Control Mechanism and Extreme Gradient
Boosting Algorithm for Protecting Software-Defined Networks Against DDoS Attacks,
IEEE Access, Vol. 8, pp. 194269-194288
An H., Ji Y., Zhang N., Hu W., Yu P., Wang Y., Jun. 2020, Dynamically Split the Traffic
in Software Defined Network Based on Deep Reinforcement Learning, in 2020 International
Wireless Communications and Mobile Computing (IWCMC), Limassol, Cyprus, pp. 806-811
Bolodurina I., Parfenov D., Jan. 2018, Comprehensive approach for optimization traffic
routing and using network resources in a virtual data center, Procedia Computer Science,
Vol. 136, pp. 62-71
Malik J., Akhunzada A., Bibi I., Imran M., Musaddiq A., Kim S. W., 2020, Hybrid Deep
Learning: An Efficient Reconnaissance and Surveillance Detection Mechanism in SDN,
IEEE Access, Vol. 8, pp. 134695-134706
Perez-Diaz J. A., Valdovinos I. A., Choo K.-K. R., Zhu D., 2020, A Flexible SDN-Based
Architecture for Identifying and Mitigating Low-Rate DDoS Attacks Using Machine Learning,
IEEE Access, Vol. 8, pp. 155859-155872
Pei J., Hong P., Xue K., Li D., Wei D. S. L., Wu F., Jun. 2020, Two-Phase Virtual
Network Function Selection and Chaining Algorithm Based on Deep Learning in SDN/NFV-Enabled
Networks, IEEE J. Select. Areas Commun., Vol. 38, No. 6, pp. 1102-1117
Witanto J. N., Lim H., 2019, Software-Defined Networking Application with Deep Deterministic
Policy Gradient, Proceedings of the 11th International Conference on Computer Modeling
and Simulation. Association for Computing Machinery, North Rockhampton, QLD, Australia,
pp. 176-179
Zhang J., Ye M., Guo Z., Yen C.-Y., Chao H. J., Oct. 2020, CFR-RL: Traffic Engineering
With Reinforcement Learning in SDN, IEEE J. Select. Areas Commun., Vol. 38, No. 10,
pp. 2249-2259
Lou K., Yang Y., Wang C., 2019, An Elephant Flow Detection Method Based on Machine
Learning, in Smart Computing and Communication, Cham, Vol. 11910, pp. 212-220
Rusek K., Suarez-Varela J., Almasan P., Barlet-Ros P., Cabellos-Aparicio A., Oct.
2020, RouteNet: Leveraging Graph Neural Networks for Network Modeling and Optimization
in SDN, IEEE J. Select. Areas Commun., Vol. 38, No. 10, pp. 2260-2270
Rusek K., Suárez-Varela J., Mestres A., Barlet-Ros P., Cabellos-Aparicio A., 2019,
Unveiling the potential of Graph Neural Networks for network modeling and optimization
in SDN, Proceedings of the 2019 ACM Symposium on SDN Research. Association for Computing
Machinery, San Jose, CA, USA, pp. 140-151
Sahoo K. S., Iqbal A., Maiti P., Sahoo B., Dec. 2018, A Machine Learning Approach
for Predicting DDoS Traffic in Software Defined Networks, in 2018 International Conference
on Information Technology (ICIT), Bhubaneswar, India, pp. 199-203
Chang L.-H., Lee T.-H., Chu H.-C., Su C.-W., Sep. 2020, Application-Based Online Traffic
Classification with Deep Learning Models on SDN Networks, Advances in Technology Innovation,
Vol. 5, No. 4
Audah M. Z. F., Chin T. S., Zulfadzli Y., Lee C. K., Rizaluddin K., 2019, Towards
Efficient and Scalable Machine Learning-Based QoS Traffic Classification in Software-Defined
Network, in Mobile Web and Intelligent Information Systems, Cham, pp. 217-229
Prasath M. K., Perumal B., 2019, A meta-heuristic Bayesian network classification
for intrusion detection, International Journal of Network Management, Vol. 29, No.
3, pp. e2047
Latah M., Toker L., Mar. 2018, A novel intelligent approach for detecting DoS flooding
attacks in software-defined networks, Int. J. Adv. Intell. Informatics, Vol. 4, No.
1, pp. 11
Latah M., Toker L., Dec. 2020, An efficient flow-based multi-level hybrid intrusion
detection system for software-defined networks, CCF Trans. Netw., Vol. 3, No. 3, pp.
261-271
Assis M. V. O., Carvalho L. F., Lloret J., Proen\c{c}a M. L., Mar. 2021, A GRU deep
learning system against attacks in software defined networks, Journal of Network and
Computer Applications, Vol. 177, pp. 102942
Raikar M. M., S m M., Mulla M. M., Shetti N. S., Karanandi M., Jan. 2020, Data Traffic
Classification in Software Defined Networks (SDN) using supervised-learning, Procedia
Computer Science, Vol. 171, pp. 2750-2759
Aibin M., Oct. 2020, LSTM for Cloud Data Centers Resource Allocation in Software-Defined
Optical Networks, in 2020 11th IEEE Annual Ubiquitous Computing, Electronics & Mobile
Communication Conference (UEMCON), New York, NY, USA, pp. 0162-0167
Ibrar M., Wang L., Muntean G.-M., Akbar A., Shah N., Malik K. R., Jan. 2021, PrePass-Flow:
A Machine Learning based technique to minimize ACL policy violation due to links failure
in hybrid SDN, Computer Networks, Vol. 184, pp. 107706
Satheesh N., et al. , Nov. 2020, Flow-based anomaly intrusion detection using machine
learning model with software defined networking for OpenFlow network, Microprocessors
and Microsystems, Vol. 79, pp. 103285
Bouacida N., Alghadhban A., Alalmaei S., Mohammed H., Shihada B., May 2017, Failure
mitigation in software defined networking employing load type prediction, in 2017
IEEE International Conference on Communications (ICC), Paris, France, pp. 1-7
Tuan N. N., Hung P. H., Nghia N. D., Van Tho N., Van Phan T., Thanh N. H., 2020, A
DDoS attack mitigation scheme in ISP networks using machine learning based on SDN,
Electronics (Switzerland), Vol. 9, No. 3, pp. 19
Rahman O., Quraishi M. A. G., Lung C.-H., Jul. 2019, DDoS Attacks Detection and Mitigation
in SDN Using Machine Learning, in 2019 IEEE World Congress on Services (SERVICES),
Milan, Italy, pp. 184-189
Sun P., Li J., Lan J., Hu Y., Lu X., Dec. 2018, RNN Deep Reinforcement Learning for
Routing Optimization, in 2018 IEEE 4th International Conference on Computer and Communications
(ICCC), Chengdu, China, pp. 285-289
Sun P., Hu Y., Lan J., Tian L., Chen M., Oct. 2019, TIDE: Time-relevant deep reinforcement
learning for routing optimization, Future Generation Computer Systems, Vol. 99, pp.
401-409
Sun P., Guo Z., Lan J., Li J., Hu Y., Baker T., May 2021, ScaleDRL: A Scalable Deep
Reinforcement Learning Approach for Traffic Engineering in SDN with Pinning Control,
Computer Networks, Vol. 190, pp. 107891
Sun P., Guo Z., Liu S., Lan J., Wang J., Hu Y., Oct. 2020, SmartFCT: Improving power-efficiency
for data center networks with deep reinforcement learning, Computer Networks, Vol.
179, pp. 107255
Wang P., Lin S.-C., Luo M., Jun. 2016, A Framework for QoS-aware Traffic Classification
Using Semi-supervised Machine Learning in SDNs, in 2016 IEEE International Conference
on Services Computing (SCC), San Francisco, CA, USA, pp. 760-765
Hadem P., Saikia D. K., Moulik S., May 2021, An SDN-based Intrusion Detection System
using SVM with Selective Logging for IP Traceback, Computer Networks, Vol. 191, pp.
108015
Zhang Q., Wang X., Lv J., Huang M., May 2020, Intelligent Content-Aware Traffic Engineering
for SDN: An AI-Driven Approach, IEEE Network, Vol. 34, No. 3, pp. 186-193
ur Rasool R., Ahmed K., Anwar Z., Wang H., Ashraf U., Rafique W., 2021, CyberPulse++:
A machine learning-based security framework for detecting link flooding attacks in
software defined networks, International Journal of Intelligent Systems
Durner R., Kellerer W., Jun. 2020, Network Function Offloading Through Classification
of Elephant Flows, IEEE Trans. Netw. Serv. Manage., Vol. 17, No. 2, pp. 807-820
Swami R., Dave M., Ranga V., 2020, Voting-based intrusion detection framework for
securing software-defined networks, Concurrency and Computation: Practice and Experience,
Vol. 32, No. 24, pp. e5927
Volkov S. S., Kurochkin I. I., Jan. 2020, Network attacks classification using Long
Short-term memory based neural networks in Software-Defined Networks, Procedia Computer
Science, Vol. 178, pp. 394-403
Garg S., Kaur K., Kumar N., Rodrigues J. J. P. C., Mar. 2019, Hybrid Deep-Learning-Based
Anomaly Detection Scheme for Suspicious Flow Detection in SDN: A Social Multimedia
Perspective, IEEE Trans. Multimedia, Vol. 21, No. 3, pp. 566-578
Mohammed S. S., et al. , Oct. 2018, A New Machine Learning-based Collaborative DDoS
Mitigation Mechanism in Software-Defined Network, in 2018 14th International Conference
on Wireless and Mobile Computing, Networking and Communications (WiMob), Limassol,
pp. 1-8
Khamaiseh S., Serra E., Xu D., Jul. 2020, vSwitchGuard: Defending OpenFlow Switches
Against Saturation Attacks, in 2020 IEEE 44th Annual Computers, Software, Applications
Conference (COMPSAC), Madrid, Spain, pp. 851-860
Gangadhar S., Sterbenz J. P. G., Sep. 2017, Machine learning aided traffic tolerance
to improve resilience for software defined networks, in 2017 9th International Workshop
on Resilient Networks Design and Modeling (RNDM), Alghero, pp. 1-7
Petrangeli S., Wu T., Wauters T., Huysegems R., Bostoen T., De Turck F., Sep. 2017,
A machine learning-based framework for preventing video freezes in HTTP adaptive streaming,
Journal of Network and Computer Applications, Vol. 94, pp. 78-92
Kumar S., Bansal G., Shekhawat V. S., Jan. 2020, A Machine Learning Approach for Traffic
Flow Provisioning in Software Defined Networks, in 2020 International Conference on
Information Networking (ICOIN), Barcelona, Spain, pp. 602-607
Abhiroop T., Babu S., Manoj B. S., Feb. 2018, A Machine Learning Approach for Detecting
DoS Attacks in SDN Switches, in 2018 Twenty Fourth National Conference on Communications
(NCC), Hyderabad, pp. 1-6
Bakhshi T., Ghita B., Oct. 2016, OpenFlow-enabled user traffic profiling in campus
software defined networks, in 2016 IEEE 12th International Conference on Wireless
and Mobile Computing, Networking and Communications (WiMob), New York, NY, pp. 1-8
Mu T.-Y., Al-Fuqaha A., Shuaib K., Sallabi F. M., Qadir J., 2018, SDN Flow Entry Management
Using Reinforcement Learning, ACM Trans. Auton. Adapt. Syst. Association for Computing
Machinery, Vol. 13, No. 2, pp. Article 11
Chin T., Xiong K., Hu C., 2018, Phishlimiter: A Phishing Detection and Mitigation
Approach Using Software-Defined Networking, IEEE Access, Vol. 6, pp. 42516-42531
Truong-Huu T., Prathap P., Mohan P. M., Gurusamy M., May 2019, Fast and Adaptive Failure
Recovery using Machine Learning in Software Defined Networks, in 2019 IEEE International
Conference on Communications Workshops (ICC Workshops), Shanghai, China, pp. 1-6
Huong T. T., Khoa N. D. D., Dung N. X., Thanh N. H., Oct. 2019, A global multipath
load-balanced routing algorithm based on Reinforcement Learning in SDN, in 2019 International
Conference on Information and Communication Technology Convergence (ICTC), Jeju Island,
Korea (South), pp. 1336-1341
Lee T.-H., Chang L.-H., Syu C.-W., Jun. 2020, Deep Learning Enabled Intrusion Detection
and Prevention System over SDN Networks, in 2020 IEEE International Conference on
Communications Workshops (ICC Workshops), Dublin, Ireland, pp. 1-6
Liu W.-X., Zhang J., Liang Z.-W., Peng L.-X., Cai J., 2018, Content Popularity Prediction
and Caching for ICN: A Deep Learning Approach With SDN, IEEE Access, Vol. 6, pp. 5075-5089
Liu W., Cai J., Chen Q. C., Wang Y., Mar. 2021, DRL-R: Deep reinforcement learning
approach for intelligent routing in software-defined data-center networks, Journal
of Network and Computer Applications, Vol. 177, pp. 102865
Sun W., Wang Z., Zhang G., Feb. 2021, A QoS-guaranteed intelligent routing mechanism
in software-defined networks, Computer Networks, Vol. 185, pp. 107709
Qin Y., Wei J., Yang W., Sep. 2019, Deep Learning Based Anomaly Detection Scheme in
Software-Defined Networking, in 2019 20th Asia-Pacific Network Operations and Management
Symposium (APNOMS), Matsue, Japan, pp. 1-4
Park Y., Kengalahalli N. V., Chang S.-Y., Nov. 2018, Distributed Security Network
Functions against Botnet Attacks in Software-defined Networks, in 2018 IEEE Conference
on Network Function Virtualization and Software Defined Networks (NFV-SDN), Verona,
Italy, pp. 1-7
Ma Z., Huang J., 2020, Research on DDoS Abnormal Traffic Detection Under SDN Network,
in Parallel Architectures, Algorithms and Programming, Singapore, pp. 368-379
Kitchenham B., Jul. 2007, Guidelines for performing Systematic Literature Reviews
in Software Engineering, Keele University, UK, EBSE Technical Report
Nassif A. B., Talib M. A., Nasir Q., Albadani H., Dakalbab F. M., 2021, Machine Learning
for Cloud Security: A Systematic Review, IEEE Access, Vol. 9, pp. 20717-20735
Nassif A. B., Talib M. A., Nasir Q., Dakalbab F. M., 2021, Machine Learning for Anomaly
Detection: A Systematic Review, IEEE Access, Vol. 9, pp. 78658-78700
Amarudin , Ferdiana R., Widyawan , 2020, A Systematic Literature Review of Intrusion
Detection System for Network Security: Research Trends, Datasets and Methods, presented
at the ICICoS 2020 - Proceeding: 4th International Conference on Informatics and Computational
Sciences
Öney M. U., Peker S., 2019, The Use of Artificial Neural Networks in Network Intrusion
Detection: A Systematic Review, presented at the 2018 International Conference on
Artificial Intelligence and Data Processing, IDAP 2018
Xie J., et al. , 2019, A Survey of Machine Learning Techniques Applied to Software
Defined Networking (SDN): Research Issues and Challenges, IEEE Commun. Surv. Tutorials,
Vol. 21, No. 1, pp. 393-430
Liu Y., Yu F. R., Li X., Ji H., Leung V. C. M., Secondquarter 2020, Blockchain and
Machine Learning for Communications and Networking Systems, IEEE Communications Surveys
Tutorials, Vol. 22, No. 2, pp. 1392-1431
Liu J., Xu Q., Mar. 2019, Machine Learning in Software Defined Network, in 2019 IEEE
3rd Information Technology, Electronic and Automation Control Conference (ITNEC),
Chengdu, China, Vol. networking, pp. 1114-1120
TextBlob , 2021, API Reference - TextBlob 0.16.0 documentation
Author
Andres J. Aparcana-Tasayco
Andres J. Aparcana-Tasayco received his Bach. Eng. in Systems Engineering in 2021
from Universidad Autonoma del Peru, Lima, Peru. His skills are Software Development,
Database Design and Programming, Machine Learning, Networking, and Network Programmability.
Additionally, he was awarded in Huawei’s ICT Competition Peru and Thesis Workshop
by Universidad Autonoma del Peru in 2021. In addition, he has published and reviewed
papers for international conferences. He was a research committee chair in his IEEE
Student Branch and is currently an IEEE, IEEE ComSoc, and IEEE Computer member. He
currently works as a research assistant at the coordination of Cybersecurity and Networking
in Instituto Nacional de Investigacion y Capacitacion de Telecomunicaciones de la
Universidad Nacional de Ingenieria, Lima, Peru. His current research interests include
virtualization technologies and Software-Defined Networking, Programmable Data Planes,
Network Monitoring, and Network Security.
Javier Gamboa-Cruzado works at the Faculty of Systems Engineering of the Universidad
Nacional Mayor de San Marcos, Lima, Peru. He is Doctor in Systems Engineering and
Doctor in Administrative Sciences. He has published several articles in inter-national
journals and conferences. His research interests are in machine learning, big data,
the internet of things, natural language processing, and business intelligence (email:
jgamboa65@hotmail.com).