Round Corner
Department of Computer and Information Science


Entity-level sentiment impact analysis in social media texts

A key aspect of sentiment analysis is identifying the target(s) of the opinion, that is, to determine which entities in a text the expressed sentiment relates to. Exploring how textual entities are related to a text’s overall sentiment can yield information on how given entities are portrayed in social media, e.g., on Twitter. This requires the application of sentiment analysis techniques as well as named entity recognition and linking, and the use of heuristic or grammatical features to determine entity relevance and sentiment strength.


Björn Gambäck Björn Gambäck
315 IT-bygget
735 93354 
NTNU logo