Classification of Security Threats in Information Systems: Sciencedirect
Classification of Security Threats in Information Systems: Sciencedirect
Classification of Security Threats in Information Systems: Sciencedirect
com
ScienceDirect
Procedia Computer Science 32 (2014) 489 – 496
Abstract
Information systems are frequently exposed to various types of threats which can cause different types of damages
that might lead to significant financial losses. Information security damages can range from small losses to entire
information system destruction. The effects of various threats vary considerably: some affect the confidentiality or
integrity of data while others affect the availability of a system. Currently, organizations are struggling to understand
what the threats to their information assets are and how to obtain the necessary means to combat them which
continues to pose a challenge. To improve our understanding of security threats, we propose a security threat
classification model which allows us to study the threats class impact instead of a threat impact as a threat varies
over time. This paper addresses different criteria of information system security risks classification and gives a
review of most threats classification models. We define a hybrid model for information system security threat
classification in order to propose a classification architecture that supports all threat classification principles and
helps organizations implement their information security strategies.
© 2014
© 2014 The Authors.
Published by Published by Elsevier
Elsevier B.V. This is B.V.
an open access article under the CC BY-NC-ND license
Selection and peer-review under responsibility of Elhadi M. Shakshuki.
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and Peer-review under responsibility of the Program Chairs.
Keywords: Information system security; Threat classification; Threat; criteria; security risk.
1877-0509 © 2014 Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license
(http://creativecommons.org/licenses/by-nc-nd/3.0/).
Selection and Peer-review under responsibility of the Program Chairs.
doi:10.1016/j.procs.2014.05.452
490 Mouna Jouini et al. / Procedia Computer Science 32 (2014) 489 – 496
1. Introduction
With the development of Information and Communication Technologies and increasing accessibility to the
Internet, organizations become vulnerable to various types of threats. In fact, their information becomes exposed to
cyber attacks and their resulting damages. Threats come from different sources, like employees’ activities or
hacker’s attacks. The financial losses caused by security breaches [4] [12] [14] [19] [20] [21] usually cannot
precisely be detected, because a significant number of losses come from smaller-scale security incidents, caused an
underestimation of information system security risk [5]. Thus, managers need to know threats that influence their
assets and identify their impact to determine what they need to do to prevent attacks by selecting appropriate
countermeasures.
Vulnerabilities consist of weaknesses in a system which can be exploited by the attackers that may lead to
dangerous impact. When vulnerabilities exist in a system, a threat may be manifested via a threat agent using a
particular penetration technique to cause undesired effects [5] [9]. The financial threat loss to organizations could
be significant. According to the 11th Annual Computer Crime and Security Survey [2], 74.3% of the total losses are
caused by: viruses, unauthorized access, laptop or mobile hardware theft and theft of proprietary information [2].
Indeed, a research conducted by McCue in [16], indicates that 70% of fraud is perpetrated by insiders rather than by
external criminals but that 90% of security controls are focused on external threats.
To find these threats, threats sources and specific areas of the system that may be affected should be known, so
the information security assets can be protected in advance [5] [9]. Thus, effective security classification is
necessary to understand and identify threats and their potential impacts. In fact, security threats can be observed and
classified in different ways by considering different criteria like source, agents, and motivations. Threats
classification helps identify and organize security threats into classes to assess and evaluate their impacts, and
develop strategies to prevent, or mitigate the impacts of threats on the system [3] [12]. There are several known
computer system attacks classifications and taxonomies in these papers [5] [6] [7] [8] [9] [10] [11]. We notice that
many investigators have proposed taxonomies that classify attacks based on the intended effect of the attack like a
denial of service attack [7] [8] [10] and others incorporate the technique by which the attacker achieves this effect,
such as bypassing authentication or authority [5] [6] [9] [11].
The paper presents a hybrid threat classification model based on combining threat classification techniques and
impacts to better identify threat’s characteristics in order to propose suitable countermeasures to reduce risks. The
rest of the paper is organized as follows. The next section outlines threat classification principles. In section 3, we
present an overview of most known information security threat classifications. In section 4, we introduce the Multi-
dimensions threat classification as a new model to classify security threats. Conclusion section ends the paper.
A taxonomy is an approximation of reality used to gain greater understanding in a field of study [1]. A literature
review [1] [13] shows the following principles for information security classification should be respected:
x Mutually exclusive: Every threat is classified in one category excludes all others because categories do not
overlap. Every specimen should fit in at most one category.
x Exhaustive: The categories in a classification must include all the possibilities (all threat specimens).
x Unambiguous: All categories must be clear and precise so that classification is certain. Every category should be
accompanied by unambiguous classification criteria defining what specimens to be placed in that category.
x Repeatable: Repeated applications result in the same classification, regardless of who is classifying.
x Accepted: All categories are logical, intuitive and practices easy to be accepted by the majority.
x Useful: It can be used to gain insight into the field of inquiry; it can be adapted to different application needs.
These principles can be used in order to evaluate threat classifications. A good tterhreat classification should
support the most presented principles [1] [4] [12] [13] [17].
Threats classifications are important because they mainly allow identifying and understanding threats
Mouna Jouini et al. / Procedia Computer Science 32 (2014) 489 – 496 491
characteristics and source to protect systems assets. Moreover, it articulates the security risks that threaten these
systems and assists in understanding the capabilities and selection of security solutions [3] [4] [5] [9] [17].
Literature review has identified several attempts of classifications. In this section we present an overview of most
commonly used information security threat classifications.
A threat is the adversary’s goal, or what an adversary might try to do to a system [7]. It is also described as the
capability of an adversary to attack a system [7]. Thus, a threat may be defined in two ways: techniques that
attackers use to exploit the vulnerabilities in your system components or impact of threats to your assets. Therefore,
we can categorize threat classification approaches into two main classes:
x Threat agent is an actor that imposes the threat on a specific asset of the system which is represented by three
classes: human, technological, and force majeure.
x Threat motivation represents the cause of the creation of the threat and it is reorganized into two classes:
deliberate and accidental threat
x Threat localization represents the origin of threats, either internal or external.
x Attackers’ prior knowledge about the system: It represents how much the attacker knows about the system in
terms of system hardware, software, employees and users knowledge.
x Criticality of the area: It represents the criticality of parts of the system which might be affected by the threat.
x Loss: It represents all losses that can occur in the system or to the organization (privacy, integrity…)
Spoofing identity, Tampering with data, Repudiation, Information disclosure, Denial of service and Elevation of
privilege. It is a goal-based approach, where an attempt is made to get inside the mind of the attacker by rating the
threats against.
Most classifications of security threat are usually limited on use of one or two criteria to classify threats and the
others presented a non exhaustive list of threats (not all threats are covered on classification) and their categories are
not mutually exclusive. This may be sufficient for stable environment (little organization) where security threats are
relatively stable, but in the constantly changing environments, organizations fail to protect against insider threats [5].
In fact, organizations are prone to several kinds of threats which affect their reputations and it is important that they
identify all threats characteristics in order to mitigate their risks.
Classification allows organization to know threats which influence their assets and the areas which each threat
could affect and hence protect their assets in advance. In addition, it helps managers to build their organizations’
information systems with less vulnerabilities [5]. In addition to that, main problems can be identified in the existing
threats work. In fact, existing classifications do not support the classification principles [1] [3] [4]. At that point, the
usual solution is to combine different classifications and create a hybrid one.
Because of the above results, we propose a hybrid model for information system security threat classification, that
we named Multi-dimensions model for threat classification intending to respect all threats classification principles.
The main idea behind our model is to combine most threats classifications criteria and show their potential impacts.
The criteria classification list obtained from the overview cited above (section 3) are:
A threat can be caused by internal, external or both external and internal entities. In this paper, we focus only on
a binary classification of the threats origin: internal or external, in order to localize the origin (or source) of a threat.
The threat agent is the actor that imposes the threat to the system. We identified three classes for our specific
classification: humans, natural disasters and technological threats. The proposed classification covers the full set of
potential agents since we include humans, chemical and physical reaction on human-made objects (technological),
and, natural for all those agents on which humans do not have any influence.
Attackers normally have a specific goal or motive for an attack on a system. These goals can cause malicious or
non malicious results.
x Malicious threats consist of inside or outside attacks caused by employees or non-employees to harm and disrupt
an organization like viruses, Trojan horses, or worms.
x Non-malicious attacks occur due to poor security policies and controls that allow vulnerabilities and errors to
take place. It is caused by ignorant employees with the aim not to harm the system.
Threat intent represents the intent of the human who caused the threat:
Mouna Jouini et al. / Procedia Computer Science 32 (2014) 489 – 496 495
x Intentional Threats: It represents threats that are result of a harmful decision. For example computer crimes, or
when someone purposely damages property or information. Computer crimes include espionage, identity theft,
child pornography, and credit card crime.
x Unintentional Threats: It represents threats that are introduced without awareness. These threats basically include
the unauthorized or accidental modification of software. Accidental error includes corruption of data caused by
programming error, user or operator error.
In our model, a security threat can cause one or several damaging impacts to systems that we divide them into
seven types: Destruction of information, Corruption of information, Theft or loss of information, Disclosure of
information, denial of use, Elevation of privilege and Illegal usage:
x Passwords change, failing to log off before leaving a workstation, careless discarding of sensitive information
are malicious accidental insider human actions
x Sabotage, data theft, data destruction and spoofing attacks are threats caused by human outsider intentional
agents. They caused malicious damage like the corruption of data.
x Wildfire, flooding, earthquakes and tidal waves are caused by accidental external natural phenomena and allow
serious impacts like destruction end corruption of data and resources.
5. Conclusion
Information security is a critical problem for individuals and organizations because it leads to great financial
losses. This work dealt with threat classification problem in order to find a generic and flexible model that allows
better understanding of the nature of threats in order to develop appropriate strategies and information security
decisions to prevent or mitigate their effects. Our model is flexible, dynamic and multidimensional and meets all
threats classification principles. However, this model is limited to a binary decomposition of the sources of threats.
The paper presented a hybrid threat classification model that allows well defining and articulating of threat
characteristics. Indeed, it serves as a guideline to determine what kind of threats influence our system and it assists
with understanding the capabilities and selection of security decisions not only by presenting threats techniques and
their potential impacts in the same model but also by combining all existing threats criteria. We envision the use of
our threat classification model to propose a Cyber Security Econometric Model and then apply it on practical
application named a cloud computing systems.
References
1. Lindqvist U, Jonsson E. How to systematically classify computer security intrusions. IEEE Symposium on Security and Privacy; 1997. 154-
163.
2. Gordon LA, Loeb MP, Lucyshyn W, Richardson R. CSI/FBI Computer Crime and Security Survey – 2006. 11th Annual CSI/FBI Computer
Crime and Security Survey; 2006.
3. Tang J, Wang D, Ming L, Li X. A Scalable Architecture for Classifying Network Security Threats. Science and Technology on Information
System Security Laboratory; 2012.
4. Howard JD. An Analysis Of Security Incidents On The Internet 1989 – 1995. Doctoral Dissertation, Carnegie Mellon University Pittsburgh,
PA, USA; 1998.
5. Geric S, Hutinski Z. Information system security threats classifications. Journal of Information and Organizational Sciences; 2007. 31: 51.
6. Chidambaram V. Threat modeling in enterprise architecture integration; 2004.
7. Swiderski F, Snyder W. Threat Modeling. Microsoft Press; 2004.
8. Meier J, Mackman A, Vasireddy S, Dunner M, Escamilla R, Murukan A. Improving we application security: threats and counter measures.
Satyam Computer Services, Microsoft Corporation; 2003.
9. Alhabeeb M, Almuhaideb A, Le P, Srinivasan B. Information Security Threats Classification Pyramid. 24th IEEE International Conference
on Advanced Information Networking and Applications Workshops: 2010. p. 208-213.
10. ISO. Information Processing Systems-Open Systems Interconnection-Basic Reference Model. Part 2: Security Architecture, ISO 7498-2;
1989.
11. Ruf L, AG C, Thorn A, GmbH A, Christen T, Zurich Financial Services AG, Gruber B, Credit Suisse AG., Portmann R, Luzer H, Threat
Modeling in Security Architecture - The Nature of Threats. ISSS Working Group on Security Architectures,
http://www.isss.ch/fileadmin/publ/agsa/ISSS-AG-Security-Architecture_Threat-Modeling_Lukas-Ruf.pdf
12. Farahmand F, Navathe SB, Sharp GP, Enslow PH. A Management Perspective on Risk of Security Threats to Information Systems,
Information Technology and Management archive; 2005;6: 202-225.
13. Amoroso EG. Fundamentals of Computer Security Technology, Prentice-Hall PTR, Upper Saddle River, NJ; 1994.
14. Shiu S, Baldwin A, Beres Y, Mont MC, Duggan G. Economic methods and decision making by security professionals. The Tenth Workshop
on the Economics of Information Security (WEIS); 2011.
15. Loch K, Carr Houston, Warkentin M. Threats to Information Systems: Today's Reality, Yesterday's Understanding, Management
Information Systems Quarterly 16.2; 1992.
16. McCue A. Beware the insider security threat, CIO Jury; 2008. http://www.silicon.com/management/cio-insights/2008/04/17/beware-
theinsider-security-threat-39188671/
17. Howard MD. LeBlanc, Writing Secure Code 2nd ed., Redmond, Washington: Microsoft Press; 2003.
18. Rasmi M, Jantan A. Attack Intention Analysis Model for Network Forensics. Software Engineering and Computer Systems; 2011. 403-411.
19. Ben Arfa Rabai L, Jouini M, Ben Aissa A, Mili A. A cybersecurity model in cloud computing environments. Journal of King Saud
University – Computer and Information Sciences; 2012; 1: 63-75.
20. Jouini M, Ben Arfa Rabai L, Ben Aissa A, Mili A. Towards quantitative measures of Information Security: A Cloud Computing case study.
International Journal of Cyber-Security and Digital Forensics (IJCSDF); 2012; 1(3): 265-279.
21. Ben Arfa Rabai L, Jouini M, Ben Aissa A, Mili A.. An economic model of security threats for cloud computing systems. International
Conference on Cyber Security, Cyber Warfare and Digital Forensic (CyberSec); 2012. 100-105.