A semantic web based core engine to efficiently perform sentiment analysis

In this paper we present a domain-independent framework that creates a sentiment analysis model by mixing Semantic Web technologies with natural language processing approaches (This work is supported by the project PRISMA SMART CITIES, funded by the Italian Ministry of Research and Education under the program PON.). Our system, called Sentilo, provides a core sentiment analysis engine which fully exploits semantics. It identifies the holder of an opinion, topics and sub-topics the opinion is referred to, and assesses the opinion trigger. Sentilo uses an OWL opinion ontology to represent all this information with an RDF graph where holders and topics are resolved on Linked Data. Anyone can plug its own opinion scoring algorithm to compute scores of opinion expressing words and come up with a combined scoring algorithm for each identified entities and the overall sentence.

Tipo Pubblicazione: 
Contributo in atti di convegno
Author or Creator: 
Reforgiato Recupero, Diego
Consoli, Sergio
Gangemi, Aldo
Nuzzolese, Andrea Giovanni
Spampinato, Daria
Springer, Berlin , Germania
ESWC 2014 Satellite Events: The Semantic Web, pp. 245–248, 2014
Resource Identifier: 
ISTC Author: 
Ritratto di Aldo Gangemi
Real name: 
Ritratto di Daria Spampinato
Real name: