Towards Scalable Defenses against Intimate Partner Infiltrations

2025-02-06

AI Summary

Intimate Partner Intrusion (IPI)—a covert form of digital violence exploiting physical proximity, interpersonal trust, and relational context—poses significant forensic and privacy challenges. Method: We propose AID, a lightweight, privacy-preserving smartphone-based system for automated IPI detection. AID introduces a novel two-stage adaptive architecture that jointly models physical proximity and relational context, fusing multimodal sensor data (touch, accelerometer, screen state) with temporal user-behavior features. It incorporates local self-calibration under differential privacy and efficient edge inference. Contribution/Results: Evaluated in a 27-participant user study, AID achieves an end-to-end Top-3 F1-score of 0.981 with only 4% false positives. It accurately distinguishes non-owner access and IPI-specific interactions, demonstrating judicial admissibility and scalability to security-clinic-grade digital forensics support.

Technology Category

• Machine Learning: Privacy
• Computer Vision: Bias, Fairness & Privacy
• Humans and AI: Interaction Techniques and Devices

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Intimate Partner Infiltration (IPI)--a type of Intimate Partner Violence (IPV) that typically requires physical access to a victim's device--is a pervasive concern in the United States, often manifesting through digital surveillance, control, and monitoring. Unlike conventional cyberattacks, IPI perpetrators leverage close proximity and personal knowledge to circumvent standard protections, underscoring the need for targeted interventions. While security clinics and other human-centered approaches effectively tailor solutions for survivors, their scalability remains constrained by resource limitations and the need for specialized counseling. In this paper, we present AID, an Automated IPI Detection system that continuously monitors for unauthorized access and suspicious behaviors on smartphones. AID employs a two-stage architecture to process multimodal signals stealthily and preserve user privacy. A brief calibration phase upon installation enables AID to adapt to each user's behavioral patterns, achieving high accuracy with minimal false alarms. Our 27-participant user study demonstrates that AID achieves highly accurate detection of non-owner access and fine-grained IPI-related activities, attaining an end-to-end top-3 F1 score of 0.981 with a false positive rate of 4%. These findings suggest that AID can serve as a forensic tool within security clinics, scaling their ability to identify IPI tactics and deliver personalized, far-reaching support to survivors.

Problem

• Automated detection of unauthorized access
• Scalable defense against digital surveillance
• Privacy-preserving monitoring for IPI activities

Innovation

• Automated IPI Detection system
• Two-stage multimodal signal processing
• Behavioral pattern calibration for accuracy