Typed Dependency Parsing pipeline for English

Description

Typed Dependency parser, trained on the on the CONLL dataset.

Dependency parsing is the task of extracting a dependency parse of a sentence that represents its grammatical structure and defines the relationships between “head” words and words, which modify those heads.

Live Demo Open in Colab Download

How to use


from sparknlp.pretrained import PretrainedPipelinein
pipeline = PretrainedPipeline('dependency_parse', lang = 'en')
annotations =  pipeline.fullAnnotate("Dependencies represents relationships betweens words in a Sentence "")[0]
annotations.keys()


val pipeline = new PretrainedPipeline("dependency_parse", lang = "en")
val result = pipeline.fullAnnotate("Dependencies represents relationships betweens words in a Sentence")(0)


nlu.load("dep.typed").predict("Dependencies represents relationships betweens words in a Sentence")

    

Results

+---------------------------------------------------------------------------------+--------------------------------------------------------+
|result                                                                           |result                                                  |
+---------------------------------------------------------------------------------+--------------------------------------------------------+
|[ROOT, Dependencies, represents, words, relationships, Sentence, Sentence, words]|[root, parataxis, nsubj, amod, nsubj, case, nsubj, flat]|
+---------------------------------------------------------------------------------+--------------------------------------------------------+

Model Information

Model Name: dependency_parse
Type: pipeline
Compatibility: Spark NLP 3.0.0+
License: Open Source
Edition: Official
Language: en

Included Models

  • DocumentAssembler
  • SentenceDetector
  • Tokenizer
  • PerceptronModel
  • DependencyParserModel
  • TypedDependencyParserModel