1. You will need to have a Java Runtime Environment (JRE) installed on your machine, version 11 at least.

2. Download the pipeline's JAR from the release page.

3. Create a project.json file:

   • follow the rest of this guide for more details, starting with the System Overview;
   • check files/project.json for an example;
   • note that in the project.json paths are relative to the location of Topic_Mapping_Pipeline.jar on your machine.

4. Run the pipeline using the following command:

$ java -jar -Xmx4g Topic_Mapping_Pipeline.jar project.json

The -Xmx option lets you adjust the maximum heap size of the application, in this instance 4 Gigabytes.

1. You will need to have a Java Development Kit (JDK) installed on your machine, version 11 at least.

2. Download the project:

   • build contains the pipeline JAR build;
   • doc contains all the user documentation (don't forget to update it as you modify, add, remove things);
   • files contains an example project.json file and sample data files to test the application;
   • lib contains all the project dependencies, see below;
   • src contains the sources.

3. Before compiling and running sources, make sure your IDE uses the correct JDK, and that you have added the dependencies to your project structure:

   • fastcsv-1.0.2.jar for reading and writing csv files;
   • json-simple-1.1.1.jar for reading and writing json files;
   • pdfbox-app-2.0.9.jar for reading pdf documents;
   • stanford-corenlp-3.9.2.jar and stanford-corenlp-3.9.2-models.jar for lemmatising text;
   • mallet.jar and mallet-deps.jar for modelling topics;
   • jbox2d-library-2.2.1.1.jar for mapping topics.

4. The main class to use for compilation and run is src/TopicMapping:

   • don't forget to instruct your IDE to take project.json as parameter;
   • you can also instruct your IDE to adjust the maximum heap size of the application using the -Xmx option.

5. When setting up the build process, to create an updated JAR of the pipeline:

   • you should instruct your IDE to extract dependencies to the target JAR file;
   • make sure to publish the new release.

6. The .idea folder and Topic_Mapping_Pipeline.iml file should allow you to set up the project with IntelliJ IDEA.

Note on using Git:

• You might need to install Git LFS to handle version control of large files, such as some JAR files.

Although it has been deprecated, the previous version of the mapping module has been written in JavaScript (js_scripts folder). If you wish to use this version, you will need to install NodeJS to execute JavaScript.