Michael Goldsmith
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Research Interests
Michael Goldsmith has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Sadie Creese
University of Oxford
5 shared publications - Ioannis Agrafiotis
University of Oxford
4 shared publications - Arnau Erola
Department of Computer Science, University of Oxford, UK
3 shared publications - Jassim Happa
University of Oxford
3 shared publications - Louise Axon
University of Oxford
3 shared publications - Alastair Janse van Rensburg
Department of Computer Science, Oxford University, 6396 Oxford, Oxfordshire United Kingdom of Great Britain and Northern Ireland
1 shared publication
Latest Publications
- Open Access
- 20 October 2023
- Published by Association for Computing Machinery (ACM) in Digital Threats: Research and Practice
- Vol. 4 (4) , 1-38
Ransomware as a Predator: Modelling the Systemic Risk to Prey
The accelerated pace with which companies, governments, and institutions embrace digital transformation is creating opportunities for economic prosperity, but also increases the threat landscape. Recent orchestrated cyber-attacks have revealed the unpredictability of the harm they can cause in our society, rendering the creation of new models that capture systemic risk more critical than ever. In this article, we model the behaviour of one of the most prominent cyber-attacks: ransomware; in particular, ransomware that propagates between organisations via the Internet. We draw concepts from epidemiological models of viral propagation to reason about policies that can reduce the systemic cyber-risk to the community. To achieve this, we present a compartment-based epidemiological model of predator-prey interactions and run simulations to validate the importance of defensive controls that reduce the propagation of ransomware. Our model suggests that with specific defensive controls in place, other response policies may also become more effective. A prey policy to not pay the ransom may improve the ability of the victim population to recover; while information-sharing may reduce the number of organisations compromised if certain conditions on the speed of threat-intelligence sharing practices are met. These results indicate the validity of the approach, which we believe could be extended to explore the impacts of a broad range of attacker and defender behaviours and characteristics of the digital environment on systemic risk.
- 20 October 2023
- Published by Association for Computing Machinery (ACM) in Digital Threats: Research and Practice
- Vol. 4 (4) , 1-38
Open AccessRansomware as a Predator: Modelling the Systemic Risk to Prey
The accelerated pace with which companies, governments, and institutions embrace digital transformation is creating opportunities for economic prosperity, but also increases the threat landscape. Recent orchestrated cyber-attacks have revealed the unpredictability of the harm they can cause in our society, rendering the creation of new models that capture systemic risk more critical than ever. In this article, we model the behaviour of one of the most prominent cyber-attacks: ransomware; in particular, ransomware that propagates between organisations via the Internet. We draw concepts from epidemiological models of viral propagation to reason about policies that can reduce the systemic cyber-risk to the community. To achieve this, we present a compartment-based epidemiological model of predator-prey interactions and run simulations to validate the importance of defensive controls that reduce the propagation of ransomware. Our model suggests that with specific defensive controls in place, other response policies may also become more effective. A prey policy to not pay the ransom may improve the ability of the victim population to recover; while information-sharing may reduce the number of organisations compromised if certain conditions on the speed of threat-intelligence sharing practices are met. These results indicate the validity of the approach, which we believe could be extended to explore the impacts of a broad range of attacker and defender behaviours and characteristics of the digital environment on systemic risk.
- Open Access
- 19 April 2022
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 67
Insider-threat detection: Lessons from deploying the CITD tool in three multinational organisations
- 19 April 2022
- Published by Elsevier in Journal of Information Security and Applications
- Vol. 67
Open AccessInsider-threat detection: Lessons from deploying the CITD tool in three multinational organisations
- 27 November 2021
- Published by Springer Nature
- p. 279-295
A Novel Behavioural Screenlogger Detection System
- 27 November 2021
- Published by Springer Nature
- p. 279-295
A Novel Behavioural Screenlogger Detection System
- 30 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. PP (15455971) , 1
Sonification to Support the Monitoring Tasks of Security Operations Centres
Sonification (the representation of data as sound) may offer a solution to some of the network-security monitoring challenges faced in security operations centres (SOCs). Prior work has shown that sonification can present network-security information to humans effectively, and indicated that security practitioners foresee potential for sonification to aid in scenarios related to their work. The use of sonification by security practitioners in tasks relevant to SOCs has not been examined, however. To address this gap, we assessed the use of sonification by security practitioners in network-security monitoring tasks in an experimental setting. We report on the results of a study in which we compared the performance of security practitioners using a Security Information and Event Management (SIEM) tool with their performance using a SIEM tool that incorporated sonification, in a primary and a non-primary monitoring task. In both tasks, a number of aspects of the monitoring performance of participants were significantly improved when sonification was used. Our results support the potential for sonification to aid in SOC tasks, and indicate a need to validate the utility of sonification systems by running them in operational SOCs.
- 30 July 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. PP (15455971) , 1
Sonification to Support the Monitoring Tasks of Security Operations Centres
Sonification (the representation of data as sound) may offer a solution to some of the network-security monitoring challenges faced in security operations centres (SOCs). Prior work has shown that sonification can present network-security information to humans effectively, and indicated that security practitioners foresee potential for sonification to aid in scenarios related to their work. The use of sonification by security practitioners in tasks relevant to SOCs has not been examined, however. To address this gap, we assessed the use of sonification by security practitioners in network-security monitoring tasks in an experimental setting. We report on the results of a study in which we compared the performance of security practitioners using a Security Information and Event Management (SIEM) tool with their performance using a SIEM tool that incorporated sonification, in a primary and a non-primary monitoring task. In both tasks, a number of aspects of the monitoring performance of participants were significantly improved when sonification was used. Our results support the potential for sonification to aid in SOC tasks, and indicate a need to validate the utility of sonification systems by running them in operational SOCs.
- Open Access
Understanding the Radical Mind: Identifying Signals to Detect Extremist Content on Twitter
The Internet and, in particular, Online Social Networks have changed the way that terrorist and extremist groups can influence and radicalise individuals. Recent reports show that the mode of operation of these groups starts by exposing a wide audience to extremist material online, before migrating them to less open online platforms for further radicalization. Thus, identifying radical content online is crucial to limit the reach and spread of the extremist narrative. In this paper, our aim is to identify measures to automatically detect radical content in social media. We identify several signals, including textual, psychological and behavioural, that together allow for the classification of radical messages. Our contribution is three-fold: (1) we analyze propaganda material published by extremist groups and create a contextual text-based model of radical content, (2) we build a model of psychological properties inferred from these material, and (3) we evaluate these models on Twitter to determine the extent to which it is possible to automatically identify online radical tweets. Our results show that radical users do exhibit distinguishable textual, psychological, and behavioural properties. We find that the psychological properties are among the most distinguishing features. Additionally, our results show that textual models using vector embedding features significantly improves the detection over TF-IDF features. We validate our approach on two experiments achieving high accuracy. Our findings can be utilized as signals for detecting online radicalization activities.
Open AccessUnderstanding the Radical Mind: Identifying Signals to Detect Extremist Content on Twitter
The Internet and, in particular, Online Social Networks have changed the way that terrorist and extremist groups can influence and radicalise individuals. Recent reports show that the mode of operation of these groups starts by exposing a wide audience to extremist material online, before migrating them to less open online platforms for further radicalization. Thus, identifying radical content online is crucial to limit the reach and spread of the extremist narrative. In this paper, our aim is to identify measures to automatically detect radical content in social media. We identify several signals, including textual, psychological and behavioural, that together allow for the classification of radical messages. Our contribution is three-fold: (1) we analyze propaganda material published by extremist groups and create a contextual text-based model of radical content, (2) we build a model of psychological properties inferred from these material, and (3) we evaluate these models on Twitter to determine the extent to which it is possible to automatically identify online radical tweets. Our results show that radical users do exhibit distinguishable textual, psychological, and behavioural properties. We find that the psychological properties are among the most distinguishing features. Additionally, our results show that textual models using vector embedding features significantly improves the detection over TF-IDF features. We validate our approach on two experiments achieving high accuracy. Our findings can be utilized as signals for detecting online radicalization activities.
- 31 March 2019
- Published by Springer Nature
- p. 111-129
A State Machine System for Insider Threat Detection
- 31 March 2019
- Published by Springer Nature
- p. 111-129
A State Machine System for Insider Threat Detection
- Open Access
- 1 January 2019
- Published by INSTICC
- p. 75-87
Nonsense Attacks on Google Assistant and Missense Attacks on Amazon Alexa
- 1 January 2019
- Published by INSTICC
- p. 75-87
Open AccessNonsense Attacks on Google Assistant and Missense Attacks on Amazon Alexa
- 20 December 2018
- Published by Audio Engineering Society in Journal of the Audio Engineering Society
- Vol. 66 (12) , 1016-1032
Sonification Mappings: Estimating Effectiveness, Polarities and Scaling in an Online Experiment
Sonification is a technique to present data arrays as sound, thereby taking advantage of the human ability to hear patterns that might otherwise not be apparent. Mappings from parameters of data to parameters of sound form the basis of parameter-mapping sonification. The choice of mappings and their design can influence both the utility of the sonification system and the ability of users to interpret the sounds. In this article the authors demonstrate the use of a time-efficient methodology...
- 20 December 2018
- Published by Audio Engineering Society in Journal of the Audio Engineering Society
- Vol. 66 (12) , 1016-1032
Sonification Mappings: Estimating Effectiveness, Polarities and Scaling in an Online Experiment
Sonification is a technique to present data arrays as sound, thereby taking advantage of the human ability to hear patterns that might otherwise not be apparent. Mappings from parameters of data to parameters of sound form the basis of parameter-mapping sonification. The choice of mappings and their design can influence both the utility of the sonification system and the ability of users to interpret the sounds. In this article the authors demonstrate the use of a time-efficient methodology...
- 23 September 2018
- Published by Springer Nature
- p. 18-32
A Survey of Keylogger and Screenlogger Attacks in the Banking Sector and Countermeasures to Them
- 23 September 2018
- Published by Springer Nature
- p. 18-32
A Survey of Keylogger and Screenlogger Attacks in the Banking Sector and Countermeasures to Them
- Open Access
- 14 August 2017
- Published by Emerald Publishing in PSU Research Review
- Vol. 1 (2) , 76-90
On properties of cyberattacks and their nuances
Purpose: Several attack models attempt to describe behaviours of attacks with the intent to understand and combat them better. However, all models are to some degree incomplete. They may lack insight about minor variations about attacks that are observed in the real world (but are not described in the model). This may lead to similar attacks being classified as the same type of attack, or in some cases the same instance of attack. The appropriate solution would be to modify the model or replace it entirely. However, doing so may be undesirable as the model may work well for most cases or time and resource constraints may factor in as well. This paper aims to explore the potential value of adding information about attacks and attackers to existing models. Design/methodology/approach: This paper investigates used cases of minor variations in attacks and how it may and may not be appropriate to communicate subtle differences in existing attack models through the use of annotations. In particular, the authors investigate commonalities across a range of existing models and identify where and how annotations may be helpful. Findings: The authors propose that nuances (of attack properties) can be appended as annotations to existing attack models. Using annotations appropriately should enable analysts and researchers to express subtle but important variations in attacks that may not fit the model currently being used. Research limitations/implications: This work only demonstrated a few simple, generic examples. In the future, the authors intend to investigate how this annotation approach can be extended further. Particularly, they intend to explore how annotations can be created computationally; the authors wish to obtain feedback from security analysts through interviews, identify where potential biases may arise and identify other real-world applications. Originality/value: The value of this paper is that the authors demonstrate how annotations may help analysts communicate and ask better questions during identification of unknown aspects of attacks faster,e.g. as a means of storing mental notes in a structured manner, especially while facing zero-day attacks when information is incomplete.
- 14 August 2017
- Published by Emerald Publishing in PSU Research Review
- Vol. 1 (2) , 76-90
Open AccessOn properties of cyberattacks and their nuances
Purpose: Several attack models attempt to describe behaviours of attacks with the intent to understand and combat them better. However, all models are to some degree incomplete. They may lack insight about minor variations about attacks that are observed in the real world (but are not described in the model). This may lead to similar attacks being classified as the same type of attack, or in some cases the same instance of attack. The appropriate solution would be to modify the model or replace it entirely. However, doing so may be undesirable as the model may work well for most cases or time and resource constraints may factor in as well. This paper aims to explore the potential value of adding information about attacks and attackers to existing models. Design/methodology/approach: This paper investigates used cases of minor variations in attacks and how it may and may not be appropriate to communicate subtle differences in existing attack models through the use of annotations. In particular, the authors investigate commonalities across a range of existing models and identify where and how annotations may be helpful. Findings: The authors propose that nuances (of attack properties) can be appended as annotations to existing attack models. Using annotations appropriately should enable analysts and researchers to express subtle but important variations in attacks that may not fit the model currently being used. Research limitations/implications: This work only demonstrated a few simple, generic examples. In the future, the authors intend to investigate how this annotation approach can be extended further. Particularly, they intend to explore how annotations can be created computationally; the authors wish to obtain feedback from security analysts through interviews, identify where potential biases may arise and identify other real-world applications. Originality/value: The value of this paper is that the authors demonstrate how annotations may help analysts communicate and ask better questions during identification of unknown aspects of attacks faster,e.g. as a means of storing mental notes in a structured manner, especially while facing zero-day attacks when information is incomplete.