Which NLP library among the following ones below is the most mature and should be used in a scaled project of text processing.
In my project I am designing a full infrastructure of unstructured text processing, and there are scenarios where none of the available libraries can be leveraged, therefore I am implementing my own solution, however, I want to make evaluation against the current state of the art NLP libraries. and I want to pick up the most mature ones.
- OpenNLP
- Natural Language Toolkit (NLTK)
- Stanford NLP
- Machine Learning for Language Toolkit (MALLET)
- LingPipe
- Freeling
- TreeTagger
Thank you in advance.
Hassan.
Well, complex question.
Depending on your requirements (research vs commercial, language coverage, tasks, scalability .... as the most relevant) I would recommend OpenNLP or Stanford.
Definitely Stanford, if you're working with English.
OpenNLP only comes trained with demonstration data, so it won't give you good results until you re-train with your own corpora.
Mr Mike Dillinger, The NER Tagger of Stanford has very poor performance even though I trained it on big training data
- Badges
- Science topic
- Similar topics
- Linguistics
- Philology
- Language