Download JAR files tagged by solr with all dependencies

Implementation of an annotation engine that links the content item to a set of possible categories from a dedicated Solr index using MoreLikeThis queries. The classification can be either applied to a complete document (text in a given language) which is the default behavior or to a specific portion of the text (using a TextAnnotation).

An in-memory EntityLinking engine that uses Lucenes FST (Finite State Transducer) technology. This engine is based on code provided by the Solr Text Tagger (https://github.com/OpenSextant/SolrTextTagger/) but provides a deep integration with Apache Stanbol (Datafile provider, NLP processing module and existing EntityLinking functionality).

This service implements the Indexer and Searcher interface of WebLab and uses a embedded SOLR engine in order to realize the functions. Unlike the other SolR service, this version does not need having SolR installed. Configuration of the SOLR service can be done through IndexerBean.xml and SearcherBean.xml whereas the SolR instance configuration resides in the solr directory.

This service implements the Indexer and Searcher interface of WebLab and connect to a remote SOLR engine in order to realize the functions. The connection is mandatory and thus the remote SOLR server should be started beforehand. Configuration of the SOLR server URL is found in src/main/webapp/WEB-INF/cxf-servlet.xml . See in particular "indexerServiceBean" and "searcherServiceBean". Moreover, the service implements several "Analyser" services that could be included in a complete search chain in order to: - enrich a ResultSet with metadata about Hits (see resultSetMetaEnricherServiceBean) - highlight snippet in ResultSet (see highlighterServiceBean) - provide facets related to the current results (see facetSuggestionServiceBean) - suggest spell correction of the original query (see spellSuggestionServiceBean)