TextToTagsAndTriples - Take a user's photo description and produce the Tags and Triples

[mantzaris]

This can be taken as a science which many do in the Semantic Web world, but we are going to focus on other elements so keep it simple.

Any word that is a Stop word is not a Tag. We will even keep words such as Adjectives to be Stop words. The Triples for the RDF will be simple (subject, object, predicate). A lot of work can be done in order to pick these out as accurately as possible but our approach relies upon 'REDUNDANCY' so it should be ok in the long run.

1. produce tags
2. extract some triple stores

*Must be javascript

1. no RDF
2. format in description [S:P:O]
3. widget user create RDF
4. user curate automated RDFs produced

[mantzaris]

the most basic form of tag entry is produced now, but the triples are not yet realized

[mantzaris]

It seems like producing triples from raw text is not something that is easily done. Apparently it is an active area of research

[mantzaris]

Taking raw text and translating it into triples (RDF) is actually an ongoing area of research and far from being completed. The scope of the feature may require a redefinition.

[MichaelHopwood]

Add double store

• [Adj] [N]
• [N] [V] [Adj]
  ... etc