Savio, We at knogee.com use the DBpedia data set locally. We use Ruby for everything, so we use the redland libraries which are C libraries, but have multiple language interfaces: http://librdf.org/
If you'd like an example of the code we use to read the .nt files and transform them into a database for the redland SPARQL queries, let me know... Cheers, Luis. Quoting Richard Cyganiak <[EMAIL PROTECTED]>: > Savio, > > On 13 Feb 2008, at 16:57, Savio Neville Spiteri wrote: >> I am doing my thesis to get my B.Sc. I.T. (Hons) degree. As part of >> my Final Year Project (which is the thesis itself) I am >> investigating how use of ontologies could be made to help users >> bookmark webpages. I would like to use dbpedia as a kind of a >> 'universal ontology' in such a way that the bookmark file structure >> would be like the 'natural order of things'. > > Interesting project. Just a word of caution: Note that DBpedia isn't > an ontology. From an ontology point of view, DBpedia is just a bunch > of instances, without a nice well-engineered class hierarchy. Two > different experimental class hierarchies are available for DBpedia > (YAGO and CWCC), but both have their flaws, are very complex, and I > wouldn't say that they represent the "natural order of things" -- > maybe rather the "natural disorder of things" ;-) > >> I have a query with regards to how I actually should interact with >> the system. Since the system would be using dbpedia over and over >> again (for example bookmarking a single page would most probably >> result in a (large) number of queries), it definitely wouldn't make >> sense to send a sparql query to the sparql endpoint each time, else >> bookmarking a single page would take very long indeed! I need to >> have a local version of the entire data set. Is this possible? > > Sure. It's quite a bit of work though. > >> (Which version of the dataset should i download from >> http://wiki.dbpedia.org/Datasets?) >> . > > The current dataset dumps are at http://wiki.dbpedia.org/Downloads . > Which dumps to choose depends of course on the data that should be > available to answer the queries. If you don't need links to images or > chinese titles, then you can skip the Images and Titles/zh dumps ... > >> And what would be the most efficient way to interact with the local >> source? > > You will need to install a triple store that supports SPARQL, then > load the dumps into the store, and then you can write your application > so that it queries your local store instead of the online DBpedia > endpoint. I can't give a recommendation for a specific triple store, > but the online DBpedia endpoint runs on OpenLink Virtuoso. > > Hope that helps, > Richard > > > >> I appreciate your help a lot, since i am a novice in this area, and >> i am still feeling a little bit like a stranger to all this! Thanks >> a lot, >> >> Thanks for your co-operation, >> >> -- >> Savio Neville Spiteri >> >> Spiteri House, Triq Papa Ġwanni Pawlu II, >> Victoria Għawdex VCT9047, Malta. >> >> t: (+356) 2155 2628 / 2720 0063 >> m: (+356) 7927 0060 >> e: [EMAIL PROTECTED] >> >> ______________________________________ >> The information in this e-mail, and any of its >> attachments, is strictly confidential and >> intended solely for the person or organisation >> to whom it is addressed. Access to this email >> by anyone else is unauthorised. If you are not >> the intended recipient, you must not copy or >> distribute it or take action in reliance on it. If >> you have received this e-mail in error, please >> notify the sender as soon as possible. >> Communication via e-mail over the internet is >> not secure and messages may be read, >> manipulated or otherwise compromised by >> third parties. Savio Neville Spiteri does not >> accept any responsibility in this eventuality. >> ______________________________________ >> ------------------------------------------------------------------------- >> This SF.net email is sponsored by: Microsoft >> Defy all challenges. Microsoft(R) Visual Studio 2008. >> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/_______________________________________________ >> Dbpedia-discussion mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion > > > ------------------------------------------------------------------------- > This SF.net email is sponsored by: Microsoft > Defy all challenges. Microsoft(R) Visual Studio 2008. > http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ > _______________________________________________ > Dbpedia-discussion mailing list > [email protected] > https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion > ------------------------------------------------------------------------- This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Dbpedia-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
