Beyond Curve Fitting: Neuro-Symbolic Agents for Context-Aware Epidemic Forecasting

2025-11-28

AI Summary

To address the challenge of modeling causal interactions between contextual factors—such as school calendars and meteorological variables—and epidemiological mechanisms in hand, foot, and mouth disease (HFMD) outbreak forecasting, this paper proposes a dual-agent neural-symbolic framework. One agent employs a large language model to parse heterogeneous contextual signals and generate semantic interpretations; the other integrates a probabilistic time-series model for high-accuracy forecasting. The core innovation lies in the neural-symbolic reasoning engine, which enables semantic disentanglement of driving factors, structured knowledge fusion, and causal inference—thereby substantially improving model interpretability and uncertainty quantification (90% prediction interval coverage: 0.85–1.00). Evaluated on real-world HFMD data, the framework achieves state-of-the-art point prediction accuracy, consistently outperforming both conventional statistical and baseline deep learning methods.

Technology Category

• Machine Learning: Neuro-Symbolic Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Symbolic Representations

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Effective surveillance of hand, foot and mouth disease (HFMD) requires forecasts accounting for epidemiological patterns and contextual drivers like school calendars and weather. While classical models and recent foundation models (e.g., Chronos, TimesFM) incorporate covariates, they often lack the semantic reasoning to interpret the causal interplay between conflicting drivers. In this work, we propose a two-agent framework decoupling contextual interpretation from probabilistic forecasting. An LLM "event interpreter" processes heterogeneous signals-including school schedules, meteorological summaries, and reports-into a scalar transmission-impact signal. A neuro-symbolic core then combines this with historical case counts to produce calibrated probabilistic forecasts. We evaluate the framework on real-world HFMD datasets from Hong Kong (2023-2024) and Lishui, China (2024). Compared to traditional and foundation-model baselines, our approach achieves competitive point forecasting accuracy while providing robust 90% prediction intervals (coverage 0.85-1.00) and human-interpretable rationales. Our results suggest that structurally integrating domain knowledge through LLMs can match state-of-the-art performance while yielding context-aware forecasts that align with public health workflows. Code is available at https://github.com/jw-chae/forecast_MED .

Problem

• Forecasting epidemics by integrating contextual drivers like school schedules and weather
• Addressing limitations of existing models in causal interpretation of conflicting factors
• Developing interpretable probabilistic forecasts for hand, foot and mouth disease surveillance

Innovation

• LLM interprets contextual signals into transmission impact
• Neuro-symbolic core combines signals with historical data
• Framework decouples interpretation from probabilistic forecasting