Heuristic Sentiment Analysis for Social Engineering Mitigation During Interactive Immersion with Smart Wearable Technology

Abstract

Distraction caused by the visual processing of multiple objects during augmented reality (AR) and other forms of interactive immersion could make users more susceptible to malicious push notifications. This risk could impact users at both the individual and organization levels as well as across industries, particularly as smart wearable devices become increasingly immersive. Stage 1 of this qualitative empirical study used a virtual presentation to simulate the user interfaces of the popular AR applications Google Lens, Google Translate, Instagram, Maps, and Pokémon GO presented to (N = 70) participants aged 18–40 who regularly used these applications. Of the two notification themes presented – familiarity and urgency – 62 of 70 participants chose the familiarity theme with which to engage. Based on these results, stage 2 of the study consulted four experts in the field of AR application development to design an artificial intelligence-equipped feature that could intercept possibly malicious artifacts entering the user’s line of sight during partial immersion in AR. This article proposes the design for a natively embedded security application configurable across all device operating system types to assess incoming content in real time. The article then draws upon the expertise of these four participants to inform a comparative analysis assessing how the heuristic sentiment analytic algorithm used by such an application compares against existing spam filter algorithms. The greatest advantage found in heuristic sentiment analysis that would improve upon existing spam filter techniques, such as Bayesian and rule-based detection, was the decreased reliance on user input. The proposed automated tool’s combination of heuristic threat recognition of emerging threats and sentiment analysis based on a pre-configured lexicon could reduce the overall time required to intercept malicious content incoming to a smart wearable device interface.

Received: 22 February 2025 | Revised: 8 April 2025 | Accepted: 23 April 2025

Conflicts of Interest

The author declares that she has no conflicts of interest to this work.

Data Availability Statement

Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Author Contribution Statement

Sarah Katz: Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing – original draft, Writing – review & editing, Visualization, Supervision, Project administration.