Predicting Tweet Posting Behavior on Citizen Security: A Hawkes Point Process Analysis

2024-02-04

arXiv.org

AI Summary

This study addresses the limitation of traditional survey methods in providing real-time, prospective assessment of public perception of safety (PoS). We propose a novel exogenous-event-driven Hawkes point process model—the first to incorporate exogenous excitation into Hawkes processes for dynamic PoS modeling—jointly capturing both the temporal self-excitation of safety-related tweets and the causal influence of real-world events (e.g., crime reports, policy announcements). By embedding exogenous covariates and applying maximum likelihood estimation, the model enables short-term high-accuracy forecasting of safety-related social media activity and interpretable attribution analysis. Experiments on real Twitter data demonstrate that our approach significantly outperforms baseline methods. Moreover, it quantifies the immediate impact intensity of diverse external events on public safety discourse, achieving both predictive foresight and mechanistic interpretability.

Technology Category

• Natural Language Processing: Safety and Robustness
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

Abstract

The Perception of Security (PoS) refers to people's opinions about security or insecurity in a place or situation. While surveys have traditionally been the primary means to capture such perceptions, they need to be improved in their ability to offer real-time monitoring or predictive insights into future security perceptions. Recent evidence suggests that social network content can provide complementary insights into quantifying these perceptions. However, the challenge of accurately predicting these perceptions, with the capacity to anticipate them, still needs to be explored. This article introduces an innovative approach to PoS within short time frames using social network data. Our model incorporates external factors that influence the publication and reposting of content related to security perceptions. Our results demonstrate that this proposed model achieves competitive predictive performance and maintains a high degree of interpretability regarding the factors influencing security perceptions. This research contributes to understanding how temporal patterns and external factors impact the anticipation of security perceptions, providing valuable insights for proactive security planning.

Problem

• Predict real-time citizen security perceptions using social media data
• Model external factors affecting security-related tweet posting behavior
• Enhance interpretability of factors influencing security perception dynamics

Innovation

• Uses Hawkes Point Process for tweet analysis
• Incorporates external factors influencing security posts
• Achieves real-time predictive performance and interpretability