Automated Framing Analysis: A Rule Based System for News Media Text

Michael STURDZA


Abstract:

The number of available digital texts grows exponentially creating a need for automated framing analysis. Using computational linguistics to identify framing functions in news articles represents a cost effective method for examining large amounts of data. The proof of concept described in this paper is a rule based system that automatically identifies framing functions present in news. The proposed system identifies the morphological, syntactical, and relational information of each word of the news article text using a parser. A set of rules is applied to the semantic role of each word enabling an automated detection of framing functions using sentiment analysis.

Keywords: Frame analysis; Automated; Semantic network; Parser; Sentiment analysis.

DOI: 10.24193/jmr.32.7

Download