Scribd Logo
Upload

Welcome to Scribd!

  • 6 views

    A Systematic Literature Review of Intrusion Detection System For Network Security: Research Trends, Datasets and Methods

    Uploaded by

    Amarudin

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    A Systematic Literature Review of Intrusion Detection System For Network Security: Research Trends, Datasets and Methods

    Uploaded by

    Amarudin
    6 views1 page

    Original Title

    1

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    6 views1 page

    A Systematic Literature Review of Intrusion Detection System For Network Security: Research Trends, Datasets and Methods

    Uploaded by

    Amarudin

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    of 1

    A Systematic Literature Review of

    Intrusion Detection System for Network Security:


    Research Trends, Datasets and Methods
    Amarudin Ridi Ferdiana Widyawan
    Department of Electrical Engineering Department of Electrical Engineering Department of Electrical Engineering
    and Information Technology and Information Technology and Information Technology
    Universitas Gadjah Mada Universitas Gadjah Mada Universitas Gadjah Mada
    Yogyakarta, Indonesia Yogyakarta, Indonesia
    [email protected] [email protected] Yogyakarta, Indonesia
    & [email protected]
    Engineering and Computer Science
    Universitas Teknokrat Indonesia
    Lampung, Indonesia
    [email protected]

    Abstract—Study on intrusion detection system (IDS) mostly system that can monitors network packets and can identify
    allow network administrators to focus on development the abnormal behavior, so it can detect malicious activities or
    activities in terms of network security and making better use of anomalous behaviors [5].
    resource. Many IDS datasets, techniques and methods
    conducted by some administrator to get a good performance of In recent years, there are several intrusion datasets and
    IDS. But, some methods, techniques and datasets published techniques that can be used to create IDS. The dataset was
    differently show that research in the field of intrusion detection divided two type, they are public dataset and private dataset.
    is losing comprehensiveness. This literature review aims to Some public dataset that used on IDS such as NSL-KDD [6],
    analyze and identify the research trends of techniques, datasets and CICIDS2017 [7]. While the private dataset that used on
    and methods used on IDS topics that published in January IDS such as GHSOM [8], and C-ADS [9]. On the other hand,
    2016 to May 2020. Based on inclusion and exclusion criteria the techniques that can be implemented in IDS to detect
    was found 62 primary studies that focus and related to IDS intrusion are using machine learning. They are such as
    topic, that focuses on seven machine learning techniques: clustering [10], classification [11], estimation [12],
    classification (81%), clustering (8%), estimation (3%), association [13], prediction [14], statistic [15] and dataset
    association (2%), prediction (2%), dataset analysis (3%) and analysis [16].
    the minor research covered only 1% for statistic. Beside that,
    the research studies used public datasets as 79% and private In a study conducted by [17], it was show that applied of
    datasets as 21%. Eighteen different methods (algorithm) have machine learning on IDS has more primary advantages
    been applied and proposed to detect intrusion. From the compared to traditional methods. Some study report that
    eighteen methods, six methods most applied in IDS, they are k applied of machine learning more using algorithms on data
    Nearest Neighbor (k-NN) 7%, Random Forest (RF) 7%, Naïve mining approach, they are J48 (Decision Tree (DT)), Support
    Bayes (NB) 15%, Decision Tree (DT) 17%, Neural Network Vector Machine (SVM), Naïve Bayes (NB), k- Nearest
    (NN) 20% and Support Vector Machine (SVM) 34%. Neighbor (k-NN), Ensemble (J48+NB), Ensemble
    Furthermore, some researchers proposed some techniques and (J48+SVM), Ensemble (J48+IBK) [18]. Other research [19]
    methods to improve the accuracy of machine learning classifier have applied the algorithms such as Deep Neural Network
    on IDS, like ensembling machine learning methods, using
    DNN), Linier Regression (LR), SVM, and NB. Based on
    boosting algorithm and combined feature selection algorithm.
    Future work may ensemble classifier methods can tackle the
    previous research, there are several research related of IDS
    classification problem and can improve accuracy in detecting can be done:
    intrusions. • Various types of datasets that used in IDS, giving
    researchers the opportunity to know which datasets are
    Keywords—Intrusion Detection System (IDS), Network commonly most used.
    Securty, Dataset, Machine Learning, Systematic Literature
    Review (SLR) • Many intrusion detection techniques are used in IDS, so
    it is necessary to know what techniques are most widely
    I. INTRODUCTION used by researchers today.
    Several studies have revealed that network security is a • Several methods have been used in IDS, but each method
    very important issue in the currently, and some intrusion has no consistent performance in detecting intrusions.
    detection systems (IDS) have been developed to protect Therefore, this study aims to conduct the (SLR) to
    network security [1]. To improve IDS performance have identify and analyze the research trends about techniques,
    been applied various machine learning, including ensemble datasets and methods used in IDS research from January
    learning, so it can be used to detect attack or intrusion [2]. 2016 to May 2020. This paper is divided into five sections.
    There are several types of attacks that can disturb in a Section I describes introduction, section II describes related
    network system, they are categorized into two categories work, section III describes methodology, section IV research
    (passive attacks and active attacks) [3]. IDS has long been results, and section V conclusion and future work.
    considered the most important reference for IDS equipment
    and implementation [4]. In other words, IDS is an intelligent

    XXX-X-XXXX-XXXX-X/XX/$XX.00 ©20XX IEEE

    You might also like