Edit on GitHub
  • John Snow Labs
  • Nov 07, 2021

BERT Sequence Classification - Detect Spam SMS

  • sms
  • spam
  • bert_for_sequence_classification
  • en
  • open_source

Description

This model is imported from Hugging Face-models. It is a BERT-Tiny version of the sms_spam dataset. It identifies if the SMS is spam or not.

  • LABEL_0 : No Spam
  • LABEL_1 : Spam

Predicted Entities

LABEL_0, LABEL_1

Download Copy S3 URI

How to use

document_assembler = DocumentAssembler() \
    .setInputCol('text') \
    .setOutputCol('document')

tokenizer = Tokenizer() \
    .setInputCols(['document']) \
    .setOutputCol('token')

sequenceClassifier = BertForSequenceClassification \
      .pretrained('bert_sequence_classifier_sms_spam', 'en') \
      .setInputCols(['token', 'document']) \
      .setOutputCol('class') \
      .setCaseSensitive(True) \
      .setMaxSentenceLength(512)

pipeline = Pipeline(stages=[document_assembler, tokenizer, sequenceClassifier])

example = spark.createDataFrame([['Camera - You are awarded a SiPix Digital Camera! call 09061221066 from landline. Delivery within 28 days.']]).toDF("text")
result = pipeline.fit(example).transform(example)

val document_assembler = DocumentAssembler()
    .setInputCol("text")
    .setOutputCol("document")

val tokenizer = Tokenizer()
    .setInputCols("document")
    .setOutputCol("token")

val tokenClassifier = BertForSequenceClassification.pretrained("bert_sequence_classifier_sms_spam", "en")
      .setInputCols("document", "token")
      .setOutputCol("class")
      .setCaseSensitive(true)
      .setMaxSentenceLength(512)

val pipeline = new Pipeline().setStages(Array(document_assembler, tokenizer, sequenceClassifier))

val example = Seq.empty["Camera - You are awarded a SiPix Digital Camera! call 09061221066 from landline. Delivery within 28 days."].toDS.toDF("text")

val result = pipeline.fit(example).transform(example)

import nlu
nlu.load("en.classify.bert.sms_spam.").predict("""Camera - You are awarded a SiPix Digital Camera! call 09061221066 from landline. Delivery within 28 days.""")

Results

['LABEL_1']

Model Information

Model Name: bert_sequence_classifier_sms_spam
Compatibility: Spark NLP 3.3.2+
License: Open Source
Edition: Official
Input Labels: [token, sentence]
Output Labels: [label]
Language: en
Case sensitive: true

Data Source

https://huggingface.co/mrm8488/bert-tiny-finetuned-sms-spam-detection

Benchmarking

   label  score
accuracy   0.98