An Artificial Intelligence–Based Framework for Automated Framing Analysis of Online Media Texts

Abstract

The increasing volume of online media content has created a growing need for artificial intelligence–based methods capable of analyzing complex semantic structures at scale. Framing analysis, which examines how texts construct meaning by emphasizing specific aspects of an issue, has traditionally relied on manual qualitative approaches that are difficult to scale and replicate. This study proposes an artificial intelligence–based framework for automated framing analysis of online media texts. The framework integrates computational text processing, semantic feature extraction, and rule-guided classification to operationalize framing as a structured analytical task. Using a corpus of 114 online news articles published between 2022 and 2025, the framework analyzes framing patterns across four analytical dimensions: problem definition, causal interpretation, moral evaluation, and treatment recommendation. The results demonstrate that the proposed framework can consistently identify dominant framing structures and recurring semantic patterns across large text collections. The analysis reveals stable distributions of framing dimensions and clear associations between semantic features and interpretive functions, indicating that framing can be detected systematically using artificial intelligence techniques. The primary value of this work lies in presenting a scalable, transparent, and interpretable framework that extends automated text analysis beyond surface-level tasks such as topic or sentiment detection. By formalizing framing analysis as an artificial intelligence–supported process, the study contributes to research in natural language processing and computational text analysis. The findings suggest that artificial intelligence–based framing analysis can support large-scale media analysis, policy monitoring, and decision-support systems that require structured interpretation of textual data.