Automatically extracting frames from media content using syntacting analysis

Wouter Van Atteveldt, Tamir Sheafer, Shaul Shenhav

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review


Content analysis of media framing is difficult to do using keyword or dictionary approaches. This paper presents an automatic Semantic Network Analysis method for extracting framing. This method consists of four steps. (1) Sentences are analysed syntactically; (2) Semantic relations are extracted from the syntactic structure; (3) actors and issues are identified in these semantic relations yielding a semantic network of relations between actors and issues; and (4) the frames are extracted from this semantic network. The method is tested by analysing the framing of the 2008-2009 Gaza war by the international news media, comparing the automatically extracted frames with a manually coded gold standard. Three of four frames had strong reliability (α > .7) and one frame had moderate reliability (α = 0.58). This paves the way for more large scale quantitative content analysis studies of framing effects and dynamics.

Original languageAmerican English
Title of host publicationProceedings of the 5th Annual ACM Web Science Conference, WebSci'13
PublisherAssociation for Computing Machinery
Number of pages8
ISBN (Print)9781450318891
StatePublished - 2013
Event3rd Annual ACM Web Science Conference, WebSci 2013 - Paris, France
Duration: 2 May 20134 May 2013

Publication series

NameProceedings of the 5th Annual ACM Web Science Conference, WebSci'13


Conference3rd Annual ACM Web Science Conference, WebSci 2013


Dive into the research topics of 'Automatically extracting frames from media content using syntacting analysis'. Together they form a unique fingerprint.

Cite this