Faceted Browsing over Wikidata triples
The system is currently running on: http://grafa.dcc.uchile.cl
- Lucene 6.5
- Tomcat 7
- RDF4J 2.2
The file facet.properties contains the following options:
- languages: list of supported languages (using the same language tag from the dataset)
- entityIRI: prefix for entities
- propertyIRI: prefix for properties
- labelIRI: predicate of the triple containing the label
- descriptionIRI: predicate of the triple containing the description
- alt_labelIRI: predicate of the triples containing alt labels or aliases
- instanceOf: IRI (with no prefix) of the type or instance of predicate
- image: IRI (with no prefix) of the image predicate
- entityPrefix: prefix of all entities (excluding the domain prefix)
The included build.xml generates several jar files. These are the most important:
- index.jar: Creates an index from all entities based on an NT file
- rank.jar: Computes the graph and the Page Rank value of each entity
- boosts.jar: Copies the index from index.jar but adding the ranks from rank.jar
- cache.jar: Generates a list of all queries with a large results set that need caching.
- instances.jar: Creates an index of all instances/types of the main index and also caches the results from the list of cache.jar
- values.jar: Creates the cache for values for every property of every query that needs caching. This process may take a couple of days.
- grafa.war: Tomcat Webapp's war file. The config file need the directory of the indexes.
- The other jar files are for debugging and generate statistics (there may also be unused classes)