Semantic Overflow

Question

2

For my project of adding semantics to the web I would like to extract information from websites (on a given domain) and my supervisor suggests I use RDFA to embed the information extracted back to the user.

I would like to be able to use a triple store to store the collected information (SESAME2) as I feel this would be a good way to store the information before processing it into a XHTML document with the RDFA added. Would it be difficult to generate RDFA embedded documents? or am I thinking totally illogically on how this should be done?

Web Crawler -> Get Semantic Info -> Triple Store -> User Searches -> Pulls RDF -> Generates RDFA Embedded page

Do you know of any IRC channels that would be suitable to discuss semantic web related things? I would like some where where I can ask questions that probably don't deserve to have a question setup here about them as my knowledge of this field is still very limited.

edited Dec 14 at 21:47

asked Dec 14 at 15:44

Malachi
45●4

2

Ask away. Don't be concerned about being flamed. – Andrew Matthews Dec 15 at 6:30

mhermans · Accepted Answer · 2009-12-14 17:30:12Z UTC

I'm not completely sure I understand your requirements, especially w.r.t. how realtime it should be and where the "User Searches"-step happens. If you provide additional info, I can edit my answer...

Anyhow, do take a look at...

... if it's realtime parsing & presenting of information: rdfquery and the Tabulator-libraries (Javascript). These will allow you to parse pages while the user is browsing, enrich those extracted triples and embed them back in the page, using RDFa if you want.
... if it more static: surf or rdfalchemy (ORM's) + a templating language of choice (Python). Should be relatively easy to write templates with the proper RDFa-syntax, and will in the values using the ORM.

For IRC, look at the Semantic Web Interest Group IRC

Extracting information from websites and places to talk

1 Answer