Abstract
Cyberstalking is generally considered to be harassment that originates online; however it is also recognised that other forms of pre-existing stalking can transfer into online environments. Cellular phones are now owned by the vast majority of adults (91% in 2010/2011) in the United Kingdom. It is believed that this number will approach, and probably reach, 95% within the near future.
Introduction
The rapid proliferation of mobile devices and social media has transformed traditional stalking into a pervasive digital threat. This paper, originally published in the International Journal of Computer Applications (2013), categorizes the technological vectors used by attackers—ranging from GPS metadata to social engineering—and proposes a risk assessment framework to help individuals and organizations mitigate these evolving dangers.
The Digital Threat Landscape
As cellular phone ownership reached near-saturation in the early 2010s, a critical gap emerged: technology was developing faster than users' ability to understand the associated risks. Features designed for convenience—such as location check-ins, photo metadata, and social networking—have been weaponized for "cyberstalking." The authors define this as the use of Information and Communication Technology (ICT) to support deliberate, repeated, and hostile behavior intended to harm others.
A primary concern is the "digital footprint." Any unique identifier left online—from daily activity logs to work locations—constitutes a footprint that stalkers can aggregate to build a comprehensive profile of a victim. This is often exacerbated by the gamification of location (e.g., Foursquare) and silent data leaks like EXIF metadata in digital photos, which embed GPS coordinates directly into shared images.
Methodology and Risk Assessment
The paper proposes a structured approach to managing these threats through Risk Assessment Modeling. Risk is defined as a function of probability (likelihood) and severity (impact).
The authors establish a four-level hierarchy of vulnerabilities, ranging from Level 1 (privileged access) to Level 4 (information gathering and reconnaissance). Furthermore, the paper suggests using information entropy and the "entropy weight method" to quantify the uncertainty and importance of various security features within a risk model, moving toward a more mathematically rigorous form of social computing security.
Key Findings
Synthesizing data from organizations like WHOA (Working to Halt Online Abuse) and UK National Statistics, the paper highlights significant trends in digital harassment.
- Gender Disparity: Data from 2000–2011 involving 3,393 incidents showed that 74% of cyberstalking victims were female, while the majority of known stalkers were male (with a significant portion remaining "unknown").
- Social Engineering: Cyberstalking often relies more on "human collaboration"—deceiving people into violating security measures—than on cracking complex technical passwords.
- Legislative Progress: The UK's Protection of Freedoms Act 2012 marked a significant milestone by creating specific offenses for stalking, providing a legal framework for prosecuting digital harassment.
Discussion and Future Directions
While the specific platforms discussed (like early versions of Facebook Places) have evolved, the underlying privacy principles remain critical. The authors call for the integration of "parental control" style logic into standard operating systems and the development of more robust anonymity tools to protect potential victims. Future risk models must account for the "enormous uncertainties" of human behavior in cybersecurity, ensuring that risk management evolves alongside the technology it seeks to secure.
Summary updated