pythainlp.parse

The pythainlp.parse module provides dependency parsing for the Thai language. Dependency parsing is a fundamental task in natural language processing that involves identifying the grammatical relationships between words in a sentence, which helps to analyze sentence structure and meaning.

Modules

dependency_parsing

pythainlp.parse.dependency_parsing(text: str, model: str | None = None, tag: str = 'str', engine: str = 'esupar') → list[list[str]] | str[source]

Parse the dependency structure of a text.

Parameters:

  • text (str) – text to be parsed

  • model (Optional[str]) – model to use with the engine (for esupar, transformers_ud, and ud_goeswith)

  • tag (str) – output type, "str" (CoNLL-U text, default) or "list"

  • engine (str) –

    engine to use for parsing. Options are:

    • esupar (default) - tokenizer, POS tagger, and dependency parser using BERT/RoBERTa/DeBERTa models. GitHub

    • spacy_thai - tokenizer, POS tagger, and dependency parser for the Thai language, using Universal Dependencies. GitHub

    • transformers_ud - TransformersUD. GitHub

    • ud_goeswith - POS tagger and dependency parser using goeswith for subwords

    • attaparse - Thai dependency parser using Stanza and PhayaThaiBERT. GitHub

Returns:

CoNLL-U text if tag is "str", otherwise a list of lists of fields

Return type:

Union[list[list[str]], str]

Raises:

NotImplementedError – if the engine is not supported

Options for model with the esupar engine:

  • th (default) - KoichiYasuoka/roberta-base-thai-spm-upos model. Hugging Face

  • KoichiYasuoka/deberta-base-thai-upos - DeBERTa(V2) model pre-trained on Thai Wikipedia texts for POS tagging and dependency parsing. Hugging Face

  • KoichiYasuoka/roberta-base-thai-syllable-upos - RoBERTa model pre-trained on Thai Wikipedia texts for POS tagging and dependency parsing (syllable level). Hugging Face

  • KoichiYasuoka/roberta-base-thai-char-upos - RoBERTa model pre-trained on Thai Wikipedia texts for POS tagging and dependency parsing (character level). Hugging Face

To train models for esupar, see the esupar GitHub repository.

Options for model with the transformers_ud engine:

  • KoichiYasuoka/deberta-base-thai-ud-head (default) - DeBERTa(V2) model pre-trained on Thai Wikipedia texts for dependency parsing (head detection using Universal Dependencies) and question answering, derived from deberta-base-thai and trained on th_blackboard.conll. Hugging Face

  • KoichiYasuoka/roberta-base-thai-spm-ud-head - RoBERTa model pre-trained on Thai Wikipedia texts for dependency parsing. Hugging Face

Options for model with the ud_goeswith engine:

  • KoichiYasuoka/deberta-base-thai-ud-goeswith (default) - DeBERTa(V2) model pre-trained on Thai Wikipedia texts for POS tagging and dependency parsing (using goeswith for subwords). Hugging Face

Example:
>>> from pythainlp.parse import dependency_parsing

>>> print(
...     dependency_parsing("ผมเป็นคนดี", engine="esupar")
... )
1       ผม      _       PRON    _       _       3       nsubj   _       SpaceAfter=No
2       เป็น     _       VERB    _       _       3       cop     _       SpaceAfter=No
3       คน      _       NOUN    _       _       0       root    _       SpaceAfter=No
4       ดี       _       VERB    _       _       3       acl     _       SpaceAfter=No

>>> print(
...     dependency_parsing("ผมเป็นคนดี", engine="spacy_thai")
... )
1       ผม              PRON    PPRS    _       2       nsubj   _       SpaceAfter=No
2       เป็น             VERB    VSTA    _       0       ROOT    _       SpaceAfter=No
3       คนดี             NOUN    NCMN    _       2       obj     _       SpaceAfter=No

The dependency_parsing function is the core component of the pythainlp.parse module. It offers dependency parsing capabilities for the Thai language. Given a Thai sentence as input, this function parses the sentence to identify the grammatical relationships between words, creating a dependency tree that represents the sentence’s structure.

Usage

To use the dependency_parsing function for Thai dependency parsing, follow these steps:

  1. Import the pythainlp.parse module.

  2. Use the dependency_parsing function with a Thai sentence as input.

  3. The function will return the dependency parsing results, which include information about the grammatical relationships between words.

Example

Here’s a basic example of how to use the dependency_parsing function:

from pythainlp.parse import dependency_parsing

# Input Thai sentence
sentence = "พี่น้องชาวบ้านกำลังเลี้ยงสตางค์ในสวน"

# Perform dependency parsing
parsing_result = dependency_parsing(sentence)

# Print the parsing result
print(parsing_result)