NLP - Cube

Sentence Splitting, Tokenization, Lemmatization, Part-of-speech Tagging, Dependency Parsing and Named Entity Recognition for more than 50 languages

Browse the code

Universal Dependencies

NLPCube is fully compatible with Universal Dependencies CONLLU format. If interested in building custom models, please consult the UD Guidelines.

Open and customizable

You have access to our entire repository, so you can easily integrate your own models and algorithms. Also you can retrain the system at any time, using custom word-embeddings and hyperparameters

State-of-the-art

We keep NLPCube updated with cutting-age Machine Learning and Natural Language Processing models. You can checkout our repository for latest models.

Roadmap

Currently we are working on multiple tasks:

Creating pre-trained models for all languages included in the Universal Dependencies Corpus
Enabling NLPCube installation as a standard PIP package
Providing a fair evaluation as part of the CONLL Shared Task initiative

Contributing

There are several ways to contribute:

Test NLPCube and help us find bugs
Notify us about state-of-the-art methods and resutls using via the GIT issue tracker
Train new models and make them available to everybody

Cite

If you use NLP-Cube in your research we would be grateful if you would cite the following paper:

NLP-Cube: End-to-End Raw Text Processing With Neural Networks, Boroș, Tiberiu and Dumitrescu, Stefan Daniel and Burtica, Ruxandra, Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, Association for Computational Linguistics. p. 171--179. October 2018

or, for the bibtex format, please click here.

@InProceedings{boro-dumitrescu-burtica:2018:K18-2,
    author    = {Boroș, Tiberiu  and  Dumitrescu, Stefan Daniel  and  Burtica, Ruxandra},
    title     = {{NLP}-Cube: End-to-End Raw Text Processing With Neural Networks},
    booktitle = {Proceedings of the {CoNLL} 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies},
    month     = {October},
    year      = {2018},
    address   = {Brussels, Belgium},
    publisher = {Association for Computational Linguistics},
    pages     = {171--179},    
    url       = {http://www.aclweb.org/anthology/K18-2017}
}