Revision: 16347
http://gate.svn.sourceforge.net/gate/?rev=16347&view=rev
Author: nirajaswani
Date: 2012-11-28 12:37:30 +0000 (Wed, 28 Nov 2012)
Log Message:
-----------
documentation for zemanta and lupedia PRs
Modified Paths:
--------------
userguide/trunk/misc-creole.tex
userguide/trunk/recent-changes.tex
Modified: userguide/trunk/misc-creole.tex
===================================================================
--- userguide/trunk/misc-creole.tex 2012-11-28 10:39:49 UTC (rev 16346)
+++ userguide/trunk/misc-creole.tex 2012-11-28 12:37:30 UTC (rev 16347)
@@ -407,7 +407,111 @@
changed by setting the removeElements parameter to false so that all recognised
chemical elements are annotated.
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\sect[sec:misc-creole:zemanta]{Zemanta Semantic Annotation Service}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+There are a number of state-of-the-art methods for semantic annotation and
+linking to DBpedia (e.g. DBpedia Spotlight, YAGO, and MusicBrainz). In
+addition, commercial web services such as AlchemyAPI, OpenCalais, and
+Zemanta are also relevant. A recent evaluation of all state-of-the-art
+LOD-based methods and tools, showed that DBpedia Spotlight and
+Zemanta have the best accuracy on annotating texts with the corresponding
+URIs from DBpedia.
+
+Zemanta API (http://developer.zemanta.com) allows application developers
+to query the Zemanta engine for contextual information about the text
+that users enter. Given a piece of text, it identifies entities in the
+text and annotates these entities with their respective URIs in the
+DBPedia. In GATE, we have provided a wrapper for the Zemanta API. This
+wrapper, internally, sends the entire document text in a number of
+batches to the Zemanta service and translates its response into GATE
+annotations. Further details on the Zemanta service can be found at
+http://developer.zemanta.com/docs/.
+
+The Zemanta Service PR can be found under the Tagger\_Zemanta plugin in
+GATE. Below, we describe the various initialization and run time parameters
+of the PR.
+
+\begin{itemize}
+\item \textbf{apiKey}: Since Zemanta is a commercial service, any
+ non-commercial usage of the service has a constraint on number of
+ requests that can be made to the Zemanta service. As on 27 November
+ 2012, this limit is set at one thousand queries per day. In order to
+ be able to use the PR, you are required to obtain such a key and
+ provide it to the PR. The key can be obtained by visiting
+ http://developer.zemanta.com/docs/ and creating an account on the
+ website.
+\item \textbf{numberOfSentencesInBatch}: Since Zemanta is a webservice,
+ only a certain size of text can be sent across for processing. The
+ number of sentences to be processed in a single batch can be
+ specified using this PR. By default, this is set to 10 sentences
+ per batch.
+\item \textbf{numberOfSentencesInContext}: Zemanta utilises contextual
+ information to identy entities and assign each of them a unique URI
+ (from DBPedia). This parameter indicates the additional number of
+ sentences to be sent, both from the left and right contexts, along
+ with the text to be disambiguated.
+\item \textbf{inputASName}: This is the annotation set where the PR looks
+ for Sentences to be processed.
+\item \textbf{outputASName}: The PR creates annotations of type
\textit{Mention}
+ for every entity it identifies in the text. Such Mention annotations are
+ then stored under the annotation set as specified by the outputASName
+ parameter.
+\end{itemize}
+
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+\sect[sec:misc-creole:lupedia]{Lupedia Semantic Annotation Service}
+%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
+
+Lupedia is a Text Enrichment Service developed by Ontotext. The service
+uses Ontotext's LKB Gazetteer to lookup words against DBpedia and
+LinkedMDB (Linked Movie Database) entities. It supports multiple languages,
+such as English, Italian and French. As part of their service, they provide
+various output filters, weights and heuristics to allow accurate matching.
+The service is aimed at performing lookup but no named entity recognition.
+Ontotext's evaluation of their lupedia API suggests that it is better than
+atleast two other similar services: AlchemyAPI and OpenCalais (see
+http://www.ontotext.com/sites/default/files/publications/lupedia-eval-results.pdf)
+for more details on their evaluation.
+
+In GATE, we have developed a wrapper around their online API. The wrapper,
+sends document content to the service and transforms response into GATE
+annotations. The wrapper is called Lupedia Service PR and can be found under
+the Tagger\_Lupedia plugin in GATE. Below, we describe various run time
+parameters of the PR.
+
+\begin{itemize}
+\item \textbf{caseSensitive}: This parameter indicates whether the lookup
+ performed against DBPedia and LinkedMDB should be case sensitive or
+ not.
+\item \textbf{datasets}: By default, the PR looks up matches of types Person,
+ Event, Place, Organisation and Work and their subtypes as defined in
+ DBPedia ontology.
+\item \textbf{keepFirstAndLongestMatch}: This heuristic allows performing
+ longest match. If set to false, it will annotate every possible match.
+\item \textbf{keepHighest}: It is possible to have multiple possible URIs for
+ a given string. If this parameter is set to true, only the one with the
+ highest score is kept and remaining low score ones are deleted.
+\item \textbf{keepSpecific}: If this parameter is set to true, only the match
+ with most specific URI is preserved.
+\item \textbf{lang}: As specified earlier, the PR supports three languages:
+ English, French and Italian. The lang parameter is to specify the
+ language of the content of the document.
+\item \textbf{outputASName}: The PR produces annotations of type Mention. The
+ annotations are stored under the annotation set with name specified
+ through this parameter.
+\item \textbf{singleGreedyMatch}: Another heuristic which affects the way
+ lookup procedure is carried out.
+\item \textbf{skipShortWords}: If set to true, this parameter ensures that
+ short words (less than 3 characters) are skipped.
+\item \textbf{skipStopWords}: If set to true, stop words are skipped during
+ the lookup procedure.
+\item \textbf{threshold}: The PR assigns every match a score. This parameter
+ specifies the minimum score for mentions to be considered as possible
+ candidates.
+\end{itemize}
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\sect[sec:misc-creole:numbers]{Annotating Numbers}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Modified: userguide/trunk/recent-changes.tex
===================================================================
--- userguide/trunk/recent-changes.tex 2012-11-28 10:39:49 UTC (rev 16346)
+++ userguide/trunk/recent-changes.tex 2012-11-28 12:37:30 UTC (rev 16347)
@@ -41,6 +41,11 @@
The TermRaider plugin (see Section~\ref{sec:creole:termraider}) has been added
to provide a toolkit and sample application for term extraction.
+Two new plugins, Tagger\_Zemanta (see Section~\ref{sec:misc-creole:zemanta})
and
+Tagger\_Lupedia (see Section~\ref{sec:misc-creole:lupedia}) have been added in
+GATE. The two plugins have PRs that wrap online annotation services provided
by
+Zemanta and Ontotext.
+
\rcSubsect{October 2012}
The PRs defined in the ANNIE plugin are now described by annotations on the
Java classes
This was sent by the SourceForge.net collaborative development platform, the
world's largest Open Source development site.
------------------------------------------------------------------------------
Keep yourself connected to Go Parallel:
INSIGHTS What's next for parallel hardware, programming and related areas?
Interviews and blogs by thought leaders keep you ahead of the curve.
http://goparallel.sourceforge.net
_______________________________________________
GATE-cvs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/gate-cvs
