Sentilo performs sentence-based sentiment analysis. It relies on FRED, a successful machine reader for the Semantic Web, and applies appropriate rules to mine FRED's graphs.

The figure on the right shows the pipeline of Sentilo. Components in dashed boxes are currently being developed.

Sentilo is the first method for opinion mining that hybridizes formal semantics, NLP and typical computational intelligence techniques.

Sentilo provides two interfaces: one to measure the "sentilometers" of a sentence, the other to visualize a semantic graph representation of a sentence enriched with opinion-related information, e.g. opinion holder, topics, sentiment scores, etc.

Sentilo is available as a REST service that returns RDF as output.

Note that you can embed Sentilo in your application calling the URL:
http://wit.istc.cnr.it/stlab-tools/sentilo/service?text=sentence&format=format
where sentence is your chosen sentence and format can be png, rdf, xml, json, turtle, N3, NT, Functional (see Sentilo's CORE for details).