sparqr.dl-learner.org

SPARQR: SPARQL Query Recommender Demo Releases of DL-Learner

Start the webapp

The tomcat7-maven-plugin were used to create this webapp. To start the webapp just give java -Xms1G -Xmx2G -jar dl-learner-standalone.jar -loggerName slf4j [--options=value]. You can see a list of possibles options by typing java -jar dl-learner-standalone.jar --help.

usage: java -jar [path to your exec war jar]
	-ajpPort <ajpPort>                     ajp port to use
	-clientAuth                            enable client authentication for https
	-D <arg>                               key=value
	-extractDirectory <extractDirectory>   path to extract war content, default value: .extract
	-h,--help                              help
	-httpPort <httpPort>                   http port to use
	-httpProtocol <httpProtocol>           http protocol to use: HTTP/1.1 or org.apache.coyote.http11.Http11Nio Protocol
	-httpsPort <httpsPort>                 https port to use
	-keyAlias <keyAlias>                   alias from keystore for ssl
	-loggerName <loggerName>               logger to use: slf4j to use slf4j bridge on top of jul
	-obfuscate <password>                  obfuscate the password and exit
	-resetExtract                          clean previous extract directory
	-serverXmlPath <serverXmlPath>         server.xml to use, optional
	-uriEncoding <uriEncoding>             connector uriEncoding default ISO-8859-1
	-X,--debug                             debug

Using the webapp

The service is available under http://localhost:8080/interfaces/rest.

To use the DL-learner you must past a configuration to the webapp. At example the command: curl --data-urlencode conf@examples/AristotlePosNeg.conf "http://127.0.0.1:8080/interfaces/rest" send the AristotlePosNeg configuration to the webapp and wait for the answer of the DL-learner.

##Examples In the folder "examples" are examples config files. All the config files uses the new SPARQL-component. The SPARQL component enables to work on data located in a triple store, the new SPARQL component improve the performance of the data recovering from triple store.

AristotlePosNeg(

curl --data-urlencode conf@examples/AristotlePosNeg.conf "http://127.0.0.1:8080/interfaces/rest"

The AristotlePosNeg example show an example of learning wit positives and negatives examples. The learning problem is based on DBpedia.

St George Dragons Players

curl --data-urlencode conf@examples/StGeorge.conf "http://nlp2rdf/interfaces/rest"

This example uses a list of rugby players of the St George Dragons and some others players. The result is a definition to retrieve all St George Dragons players.