All Peregrine libraries can be fetched as Maven modules. The ​Maven Getting Started Guide provides a good introduction to Maven. Here you can find a sample property configuration file and a sample client code using the Maven artifacts. [Freek: the page with the sample property configuration file mentions the Spring configuration file. Is Spring needed for this example or is Maven sufficient?]

Building the Peregrine modules

All Peregrine modules can be build locally by checking out and building the projects. This is a three step process.

1. Checkout the source code

   svn checkout https://trac.nbic.nl/svn/data-mining/trunk peregrine
   

2. Build the third-party modules

   cd peregrine/3rd-party/lvg
   mvn clean install
   

3. Build Peregrine

   cd ../../data-mining/
   mvn clean install
   

From version 1.3 onwards Peregrine stable artifacts will be published in maven central.

pom.xml configuration

Insert the following dependencies into your project pom.xml to use Peregrine.

<dependency>
	<groupId>org.erasmusmc.data-mining.peregrine</groupId>
	<artifactId>peregrine-api</artifactId>
	<version>1.1-SNAPSHOT</version>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.ontology</groupId>
	<artifactId>ontology-api</artifactId>
	<version>1.1-SNAPSHOT</version>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.peregrine</groupId>
	<artifactId>peregrine-normalizer</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.peregrine</groupId>
	<artifactId>peregrine-tokenizer</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.peregrine</groupId>
	<artifactId>peregrine-disambiguator</artifactId>
	<version>1.1-SNAPSHOT</version>
        <scope>runtime</scope>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.peregrine</groupId>
	<artifactId>peregrine-impl-hash</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>
<!-- using DB ontology -->
<dependency>
	<groupId>org.erasmusmc.data-mining.ontology</groupId>
	<artifactId>ontology-impl-db</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>
<!-- using File ontology -->
<dependency>
	<groupId>org.erasmusmc.data-mining.ontology</groupId>
	<artifactId>ontology-impl-file</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>
<dependency>
	<groupId>org.erasmusmc.data-mining.ontology</groupId>
	<artifactId>ontology-impl-cache</artifactId>
	<version>1.1-SNAPSHOT</version>
	<scope>runtime</scope>
</dependency>

Spring configuration

Peregrine is developed, based on the ​spring application framework. Hence it is easy to incorporate with your new / existing spring applications.

Here is an example spring configuration snippet to incorporate peregrine running with a single file ontology into your spring project.

<import resource="classpath:/org/erasmusmc/data_mining/ontology/impl/file/ontology-file.context.xml" />
<alias name="singleFileOntology" alias="ontology" />
<import resource="classpath:/org/erasmusmc/data_mining/peregrine/tokenizer/impl/tokenizer.context.xml" />
<import resource="classpath:/org/erasmusmc/data_mining/peregrine/normalizer/impl/normalizer.context.xml" />
<import resource="classpath:/org/erasmusmc/data_mining/peregrine/disambiguator/impl/rule_based/disambiguator-complete.context.xml" />
<import resource="classpath:/org/erasmusmc/data_mining/peregrine/disambiguator/impl/disambiguation-decision-maker-complete.context.xml" />
<import resource="classpath:/org/erasmusmc/data_mining/peregrine/impl/hash/peregrine.context.xml" />

Notes

• lvg normalizer is packaged as a Maven artifact. You will have to set the correct value for "LVG_DIR" in lvg2006lite/data/config/lvg.properties.
• ontology file: you can get a sample ontology file from the Download section.