Servlet Configuration

This page explains the SPARQL configuration options available. The default configuration is located at recommendation-servlet/src/main/resources/default-config.xml.

---

Configure your data endpoint

You can configure the SPARQL endpoint where your data is loaded using the proxy XML element.

• backend: specify the type of the SPARQL endpoint used, i.e., HTTP, NATIVE, RDBMS, MEMORY;

• backendArgs*: a list of arguments to be passed to the backend implementation, e.g., the URL of an endpoint in case of HTTP.

    ...
    <proxy>
        <backend>HTTP</backend>
        <backendArgs>http://path/to/endpoint/</backendArgs>
    </proxy>
    ...
  

Configure the summary endpoint

You can configure the SPARQL endpoint where the summary computed over your data is loaded using the recommender XML element.

• backend: specify the type of the SPARQL endpoint used, i.e., HTTP, NATIVE, RDBMS, MEMORY;

• backendArgs*: a list of arguments to be passed to the backend implementation, e.g., the URL of an endpoint in case of HTTP;

• pagination: get solutions to the summary query by slices of X. If X == 0, the pagination is disabled, i.e., all results are returned with one query only;

• limit: add a LIMIT clause with a value X to the summary query. If X == 0, there is no limit;

• classAttributes*: set the list of predicates that define an object as a class;

• graphSummaryGraph: set the GRAPH name of the summary. By default, it is http://sindice.com/analytics.

    ...
    <recommender>
        <backend>HTTP</backend>
        <backendArgs>http://path/to/endpoint/</backendArgs> <!-- Excepts for NATIVE, this can be the same as the one used in __proxy__ -->
        <pagination>1000</pagination>
        <limit>0</limit>
        <classAttributes>http://www.w3.org/1999/02/22-rdf-syntax-ns#type</classAttributes>
        <classAttributes>http://purl.org/dc/elements/1.1/type</classAttributes>
        <graphSummaryGraph>http://summary.graph.name</graphSummaryGraph>
    </recommender>
    ...