On Textual Analysis and Machine Learning for Cyberstalking Detection

Frommholz, Ingo; Al-Khateeb, Haider; Potthast, Martin; Ghasem, Zinnar; Shukla, Mitul; Short, Emma

doi:10.1007/s13222-016-0221-x

Cited by 37 publications

(14 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Author identification is an important task to detect or reveal the culprit in terms of cybercrime and cyber-attacks [14]. As stated by [15], Author Identification task involves techniques in performing forensics of online messages to collect practical evidence by automatically analyses a large collection of suspicious online messages from a number of suspects.…”

Section: State-of-art Author Identificationmentioning

confidence: 99%

Author identification for Under-Resourced language (KadazanDusun)

Tarmizi

Saee

Ibrahim

2020

IJEECS

View full text Add to dashboard Cite

<span>This paper presents the task of Author Identification for KadazanDusun language by using tweets as the source of data to perform Author Identification task of short text on KadazanDusun, which is considered as one the under-resourced language in Malaysia. The aim of this paper is to demonstrate Author Identification of short text on KadazanDusun. Besides, this paper also examines the performance of two machine learning algorithms on the KadazanDusun data set by analyzing the stylometric features. Stylometric features are used to quantify the writing styles of the authors which includes character n-grams and word n-grams. The workflow of Author Identification implements the machine learning approach to solve the single-labelled multi-class problem and predict the author of a given message in KadazanDusun. Two classifiers are used to compare the accuracy including Naïve Bayes and Support Vector Machine (SVM). The results show that the combination of n-grams which is word-level unigram and {1-5}-grams with character 3-grams are the most relevant stylometric features in identifying the author of KadazanDusun message with an accuracy of 80.17%. The results also show that SVM classifier has outperformed Naive Bayes in this Author Identification task with the accuracy of 80.17%.</span>

show abstract

Section: State-of-art Author Identificationmentioning

confidence: 99%

Author identification for Under-Resourced language (KadazanDusun)

Tarmizi

Saee

Ibrahim

2020

IJEECS

View full text Add to dashboard Cite

show abstract

“…Behaviour of like-minded people and drivers in similar roles can be argued to behave in almost identical patterns. Machine learning have been utilised to detect different types of online perpetrators based on their social interactions in the cyberspace [33]. Likewise, Bus drivers could have very similar reactions to speed, route and alertness.…”

Section: Behavioural Profilingmentioning

confidence: 99%

Proactive Threat Detection for Connected Cars Using Recursive Bayesian Estimation

et al. 2018

View full text Add to dashboard Cite

Abstract-Upcoming disruptive technologies around autonomous driving of connected cars have not yet been matched with appropriate security by design principles and lack approaches to incorporate proactive preventative measures in the wake of increased cyber-threats against such systems. In this paper, we introduce proactive anomaly detection to a use-case of hijacked connected cars to improve cyber-resilience. Firstly, we manifest the opportunity of behavioural profiling for connected cars from recent literature covering related underpinning technologies. Then, we design and utilise a new dataset file for connected cars influenced by the Automatic Dependent Surveillance -Broadcast (ADS-B) surveillance technology used in the aerospace industry to facilitate data collection and sharing. Finally, we simulate the analysis of travel routes in real-time to predict anomalies using predictive modelling. Simulations show the applicability of a Bayesian estimation technique, namely Kalman Filter. With the analysis of future state predictions based on the previous behaviour, cyber-threats can be addressed with a vastly increased time-window for a reaction when encountering anomalies. We discuss that detecting real-time deviations for malicious intent with predictive profiling and behavioural algorithms can be superior in effectiveness than the retrospective comparison of known-good/known-bad behaviour. When quicker action can be taken while connected cars encounter cyber-attacks, more effective engagement or interception of command and control will be achieved.

show abstract

“…Alternatively, new laws have emerged that consider cyberstalking to be a criminal offence. (Hazelwood and Koon-Magnin 2013) Intervention and defence strategies against cyberstalking can be facilitated further by means of developing technology to automatically detect, classify, filter and consequently block unwanted messages (Ghasem, Frommholz et al 2015, Frommholz, al-Khateeb et al 2016. Research shows virtual agents have been proposed to provide advice and social support (al-Khateeb and Epiphaniou 2016).…”

Section: Intervention and Strategies Against Cyberstalkingmentioning

confidence: 99%

Cyberstalking: Investigating formal intervention and the role of Corporate Social Responsibility

Al-Khateeb

Epiphaniou

Alhaboby

et al. 2017

Telematics and Informatics

Self Cite

View full text Add to dashboard Cite

Abstract-Context: Online harassment and stalking have been identified with growing accordance as anti-social behaviours, potentially with extreme consequences including indirect or direct physical injury, emotional distress and/or financial loss. Objective: As part of our ongoing work to research and establish better understanding of cyberstalking, this study aims to investigate the role of Police, Mobile Operators, Internet Service Providers (ISPs) and owners/administrators of online platforms (e.g. websites, chatrooms) in terms of intervention in response to offences. We ask to what different authorities do people report incidents of cyberstalking? Do these authorities provide satisfactory responses or interventions? And how can this be improved? Furthermore, we discuss the role of Corporate Social Responsibility (CSR) to encourage the implementation of cyberstalking-aware schemes by service providers to support victims. In addition, CSR can be used as a means to measure the effects of externality factor in dictating the relationship between the impact of a given individuals' privacy loss and strategic decisions on investment to security controls in an organisational context. Method: A Mixed design method has been used in this study. Data collection took place by means of an online survey made available for three years to record both qualitative and quantitative data. Overall, 305 participants responded from which 274 identified themselves as victims of online harassment. Result: Our results suggest that most offences were communicated through private channels such as emails and/or mobile texts/calls. A significant number of victims did not report this to their service provider because they did not know they could. While Police were recognised as the first-point-of-contact in such cases, 41.6% of our sample did not contact the Police due to reasons such as fear of escalation, guilt/sympathy and self-blaming. Experiences from those who have reported offences to service providers demonstrate that no or very little support was offered. Overall, the majority of participants shared the view that third-party intervention is required on their behalf in order to mitigate risks associated with cyberstalking. An independent specialist antistalking organisation was a popular choice to act on their behalf followed by the Police and network providers. Conclusion: Incidents are taking place on channels owned and controlled by large, cross-border international companies providing mobile services, webmail and social networking. The lack of support offered to victims in many cases of cyberstalking can be identified as Corporate Social Irresponsibility (CSI). We anticipate that awareness should be raised as regarding service providers' liability and social responsibility towards adopting better strategies.

show abstract

On Textual Analysis and Machine Learning for Cyberstalking Detection

Cited by 37 publications

References 22 publications

Author identification for Under-Resourced language (KadazanDusun)

Author identification for Under-Resourced language (KadazanDusun)

Proactive Threat Detection for Connected Cars Using Recursive Bayesian Estimation

Cyberstalking: Investigating formal intervention and the role of Corporate Social Responsibility

Contact Info

Product

Resources

About