Stop Words Cleaner for French

Description

This model removes ‘stop words’ from text. Stop words are words so common that they can be removed without significantly altering the meaning of a text. Removing stop words is useful when one wants to deal with only the most semantically important words in a text, and ignore words that are rarely semantically relevant, such as articles and prepositions.

Open in Colab Download

How to use

...
stop_words = StopWordsCleaner.pretrained("stopwords_fr", "fr") \
        .setInputCols(["token"]) \
        .setOutputCol("cleanTokens")
nlp_pipeline = Pipeline(stages=[document_assembler, tokenizer, stop_words])
light_pipeline = LightPipeline(nlp_pipeline.fit(spark.createDataFrame([['']]).toDF("text")))
results = light_pipeline.fullAnnotate("En plus d'être le roi du nord, John Snow est un médecin anglais et un leader dans le développement de l'anesthésie et de l'hygiène médicale.")
...
val stopWords = StopWordsCleaner.pretrained("stopwords_fr", "fr")
        .setInputCols(Array("token"))
        .setOutputCol("cleanTokens")
val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, stopWords))
val result = pipeline.fit(Seq.empty["En plus d'être le roi du nord, John Snow est un médecin anglais et un leader dans le développement de l'anesthésie et de l'hygiène médicale."].toDS.toDF("text")).transform(data)
import nlu

text = ["""En plus d'être le roi du nord, John Snow est un médecin anglais et un leader dans le développement de l'anesthésie et de l'hygiène médicale."""]
stopword_df = nlu.load('fr.stopwords').predict(text)
stopword_df[["cleanTokens"]]

Results

[Row(annotatorType='token', begin=8, end=13, result="d'être", metadata={'sentence': '0'}),
Row(annotatorType='token', begin=18, end=20, result='roi', metadata={'sentence': '0'}),
Row(annotatorType='token', begin=25, end=28, result='nord', metadata={'sentence': '0'}),
Row(annotatorType='token', begin=29, end=29, result=',', metadata={'sentence': '0'}),
Row(annotatorType='token', begin=31, end=34, result='John', metadata={'sentence': '0'}),
...]

Model Information

Model Name: stopwords_fr
Type: stopwords
Compatibility: Spark NLP 2.5.4+
Edition: Official
Input Labels: [token]
Output Labels: [cleanTokens]
Language: fr
Case sensitive: false
License: Open Source

Data Source

The model is imported from https://github.com/WorldBrain/remove-stopwords