From: Pablo M. <pab...@gm...> - 2011-02-15 15:14:50
|
Hi Rocco, Thanks for your interest and for suggesting improvements. You are right, co-reference resolution was not a focus of this first release. At this point we only do some simple rule-based propagation of references. It is within our interests for future work, however, to use machine learning to improve our co-reference resolution. If you are interested in contributing towards that, please feel free to contact us. At this stage we focused more on the flexibility, in order to allow people to reuse the software in many use cases. If you want to reduce the risk of getting many mismatches, you can always configure the confidence parameter to a higher value. I can see, for example, that with confidence 0.6 we only see one error for the text you provided. See: http://bit.ly/gYwNaG Best regards, Pablo Mendes ---------- Forwarded message ---------- > From: Rocco Tripodi <roc...@gm...> > Date: Tue, Feb 15, 2011 at 14:29 > Subject: Re: [Dbpedia-discussion] Dbpedia-discussion Digest, Vol 48, Issue > 4 > To: dbp...@li... > > > Hello, > I am working on the same project at the Ca' Foscari University of > Venice. In my opinion you should improve the coreference resolution > module. In fact in sentences like the following you can find many > mismatches > > > The Wizards may not want Jordan, but the head of the expansion team says > Jordan could run operations there, if he wants to. It's his home state after > all. On Wednesday, Washington Wizards owner ended the three year association > with Jordan. According to the Washington Post , the meeting degenerated into > name calling after Jordan was offered $10 million in gratitude over the next > five years. Jordan reportedly responded he was not in it for the money. > > Cnn's Josie Karp has the story. For the first time since being cut from > the team in high school, Michael Jordan was left in the awkward position of > being unwanted by the game he has come to define. After playing the last two > seasons with the Washington Wizards, Jordan said he expected to return to > the front office. Instead, during a meeting that lasted only minutes, team > owner told Jordan his services were no longer necessary. Jordan issued a > statement expressing his shock over. > > > Regards > R cc Tr p d > > > > > > Message: 3 > > Date: Mon, 14 Feb 2011 19:41:51 +0100 > > From: Max Jakob <max...@fu...> > > Subject: [Dbpedia-discussion] [ANN] DBpedia Spotlight - Text > > Annotation with DBpedia Resources > > To: dbpedia-discussion <dbp...@li...> > > Message-ID: > > <AAN...@ma...> > > Content-Type: text/plain; charset=UTF-8 > > > > Hi all, > > > > we are happy to announce a first release of DBpedia Spotlight - > > Shedding Light on the Web of Documents. > > > > The amount of data in the Linked Open Data cloud is steadily > > increasing. Interlinking text documents with this data enables the Web > > of Data to be used as background knowledge within document-oriented > > applications such as search and faceted browsing. > > > > DBpedia Spotlight is a tool for annotating mentions of DBpedia > > resources in text, providing a solution for linking unstructured > > information sources to the Linked Open Data cloud through DBpedia. The > > DBpedia Spotlight Architecture is composed by the following modules: > > ?? ?* Web application, a demonstration client (HTML/Javascript UI) > > that allows users to enter/paste text into a Web browser and visualize > > the resulting annotated text. > > ?? ?* Web Service, a RESTful Web API that exposes the functionality of > > annotating and/or disambiguating entities in text. The service returns > > XML, JSON or RDF. > > ?? ?* Annotation Java / Scala API, exposing the underlying logic that > > performs the annotation/disambiguation. > > ?? ?* Indexing Java / Scala API, executing the data processing > > necessary to enable the annotation/disambiguation algorithms used. > > > > More information about DBpedia Spotlight can be found at: > > > > http://spotlight.dbpedia.org > > > > DBpedia Spotlight is provided under the terms of the Apache License, > > Version 2.0. Part of the code uses LingPipe under the Royalty Free > > License. The source code can be downloaded from: > > > > http://sourceforge.net/projects/dbp-spotlight > > > > The development of DBpedia Spotlight was supported by: > > > > ?? ?* Neofonie GmbH, a Berlin-based company offering leading > > technologies in the area of Web search, social media and mobile > > applications (http://www.neofonie.de/). > > ?? ?* The European Commission through the project LOD2 ? Creating > > Knowledge out of Linked Data (http://lod2.eu/). > > > > Lots of thanks to: > > ?? ?* Andreas Schultz for his help with the SPARQL endpoint. > > ?? ?* Paul Kreis for his help with evaluations. > > ?? ?* Robert Isele and Anja Jentzsch for their help in early stages > > with the DBpedia extraction framework. > > > > Cheers, > > Pablo N. Mendes, Max Jakob, Andr?s Garc?a-Silva and Chris Bizer. > > > > > > > > |