Roberto Mirizzi wrote: > Il 06/05/2010 13.11, Kingsley Idehen ha scritto: >>> BTW, that would be great if the DBpedia project could compute and >>> distribute the PageRank or the TunkRank [1] values for the DBpedia >>> resources based on the data of the page links graph. This is a really >>> good scoring heuristic when performing fuzzy text named queries with >>> several homonymic matches. >>> >>> [1] http://thenoisychannel.com/2009/01/13/a-twitter-analog-to-pagerank/ >>> >>> >> Olivier, >> >> Have you looked at this interface to DBpedia: http://dbpedia.org/fct ? >> Also look at the Entity Rank details in the "About" section. >> >> Basically, you have two ranking schemes in place: >> >> 1. Entity Rank -- based on Link Coefficients >> 2. Text Scores >> > > In [1] we compute the similarity value among pairs of resources by > applying an hybrid approach to rank node exploiting both the graph > structure of the underlying RDF DBpedia graph and querying several > external information sources (search engines and social tagging > system). I think it can be useful when you need to know resources that > are more relevant than other w.r.t. a given "query". > > [1] R. Mirizzi, A. Ragone, T. Di Noia, E. Di Sciascio. Ranking the > Linked Data: the case of DBpedia. 10th International Conference on Web > Engineering (ICWE). 2010. > http://sisinflab.poliba.it/publications/2010/MRDD10b/ >
How do you pull in the external data sources? If you enable owl:sameAs and/or IFP based inference re. Virtuoso you will get a union expansion across all the Named Graphs in the instance. If you add Sponger Cartridge pragmas to the mix, it will crawl (you can designate paths). Post crawl it can even include Transitivity based path expressions, and after all of that still apply Entity Rank and Text Scores. If all else fails for specific needs, you can simply slot in your own Entity Ranking algorithm, its that flexible :-) -- Regards, Kingsley Idehen President & CEO OpenLink Software Web: http://www.openlinksw.com Weblog: http://www.openlinksw.com/blog/~kidehen Twitter/Identi.ca: kidehen ------------------------------------------------------------------------------ _______________________________________________ Dbpedia-discussion mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dbpedia-discussion
