Laboratoire ICAR » Plateforme TXM

Solution spec¶

• create a general NLP interface/component
• add a tokenizer, tagger, lemmatizer, parser, etc. sub-interfaces to the NLP component
• add a setProcessor method to the tagger interface
• add a language interface to the tagger interface
• add a addLanguage method to the tagger interface
• add a getAvailableLanguages method to the language interface
• add a new NLP preferences UI
• add a setTaggerProcessor UI in the NLP preferences UI
• add a tagger preferences UI (depends on the processor)
• Solution 1
  • populate a availableLanguages field of the tagger interface with the addLanguage method from the processor parameters directory
• Solution 2
  • add a Add Language UI in the tagger preferences UI
  • populate a availableLanguages field of the tagger interface with the addLanguage method from the Add Language UI
• etc.
• build the "Main language" input parameter values list by calling the getAvailableLanguages method

Solution implementation¶

Check available NLP interfaces before building a new one: eg. (sorted by interest) DKPro, Stanford NLP, OpenNLP, LingPipe, Lucene and Solr, GATE , UIMA, MALLET, Icarus

Check available NLP UI before building a new one: eg. WebLicht, TextFlows