Hi!
Is there any Java open-source software, which can determine words normal forms. I need to develop grammatic class for Apache Solr for Latvian language. My research showed, that there is not any class for latvian grammar for Apache Solr or Apache Lucene, so, i need to develop it by myself.
I thought, that it would be convenient, if i transform all given text in normal forms, and then index it. For example:
I have such a sentence: "Jack`s life wasn't full of romantic moments." In normal forms it is: "Jack life is not full of romantic moment". In latvian this sentence is quite more complex with grammar, so algorithm is also more complex, than in English.
I tried Hunspell, but it can't give me normal forms, just correct my grammar.
Maybe there is some other developments? Hunspell library itself can create from word its normal form, but with jar given, i can`t do it.