swarmauri_parser_keywordextractor 0.11.0.dev1

Keyword Extractor Parser for Swarmauri.

Swarmauri Parser Keywordextractor

KeywordExtractorParser wraps the YAKE keyword extraction library to turn arbitrary text into a ranked list of swarmauri_standard.documents.Document instances. Each returned document stores the detected keyword in content and the YAKE importance score in metadata["score"].

The parser normalizes any input into a string before analysis and, by default, extracts up to 10 keywords using the English language model, three-word maximum phrases, and YAKE's sequence-matching deduplication (dedupLim=0.9). Override lang or num_keywords when instantiating the parser to tailor the output to your dataset.

Installation

Choose the tool that matches your workflow:

# pip
pip install swarmauri_parser_keywordextractor

# Poetry
poetry add swarmauri_parser_keywordextractor

# uv
uv add swarmauri_parser_keywordextractor

Usage

Here's a basic example of how to use the KeywordExtractorParser:

from swarmauri_parser_keywordextractor import KeywordExtractorParser

# Initialize the parser for three keywords in English
parser = KeywordExtractorParser(num_keywords=3, lang="en")

text = "Artificial intelligence and machine learning are transforming technology"
documents = parser.parse(text)

for document in documents:
    score = document.metadata["score"]
    print(f"Keyword: {document.content}, Score: {score:.4f}")

Each call to parse returns a list of Document objects ranked by YAKE so you can feed them directly into downstream Swarmauri pipelines.

Want to help?

If you want to contribute to swarmauri-sdk, read up on our guidelines for contributing that will help you get started.