# SparkNLP POS

**Supported Labeling Types**: `Span labeling`

SparkNLP Part-of-Speech (POS) Tagging is a fast and scalable component of the SparkNLP library that assigns grammatical tags—such as noun, verb, adjective, or adverb—to each word in a sentence. It uses advanced NLP models optimized for large-scale processing, making it suitable for handling massive datasets efficiently. In our labeling platform, SparkNLP POS tagging enhances text analysis by providing syntactic insights that can improve label suggestions, rule-based automation, and overall annotation quality.

<figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-ee7040f86512d54de783015a832700925cae6cc9%2FExtension%20-%20ML-assisted%20Labeling%20-%20Span%20labeling%20-%20SparkNLP%20POS%20-%20highlight.png?alt=media" alt="Image of ML Assisted with SparkNLP POS"><figcaption><p>ML Assisted with SparkNLP POS</p></figcaption></figure>

### Model Details

* POS-tagging in SparkNLP is done via the `en.pos` model from [johnsnowlabs/nlp\_server](https://nlp.johnsnowlabs.com/docs/en/nlp_server/nlp_server).
* Models are trained primarily on the Penn Treebank corpus, supplemented with diverse web content to improve robustness across text types.
* Operates as a service accessible within the Datasaur Intelligence container.

### Usage

* SparkNLP POS tagging is ideal for large-scale text processing, including syntactic analysis and document parsing.
* The tagset is **similar to the** [**NLTK** ](https://docs.datasaur.ai/assisted-labeling/ml-assisted-labeling/sparknlp-pos)**provider**.

### References

* <https://nlp.johnsnowlabs.com/docs/en/nlp_server/nlp_server>
* <https://nlu.johnsnowlabs.com/docs/en/examples#part-of-speech--pos>