Description
This model is a BioBERT based sentiment analysis model that can extract information from COVID-19 pandemic-related tweets. The model predicts whether a tweet contains positive, negative, or neutral sentiments about COVID-19 pandemic.
Predicted Entities
neutral, positive, negative
How to use
document_assembler = DocumentAssembler() \
.setInputCol("text") \
.setOutputCol("document")
tokenizer = Tokenizer() \
.setInputCols(["document"]) \
.setOutputCol("token")
sequenceClassifier = MedicalBertForSequenceClassification.pretrained("bert_sequence_classifier_covid_sentiment", "en", "clinical/models")\
.setInputCols(["document","token"])\
.setOutputCol("class")
pipeline = Pipeline(stages=[
document_assembler,
tokenizer,
sequenceClassifier
])
data = spark.createDataFrame([
["British Department of Health confirms first two cases of in UK"],
["so my trip to visit my australian exchange student just got canceled bc of coronavirus. im heartbroken :("],
[ "I wish everyone to be safe at home and stop pandemic"]]
).toDF("text")
result = pipeline.fit(data).transform(data)
result.select("text", "class.result").show(truncate=False)
val documenter = new DocumentAssembler()
.setInputCol("text")
.setOutputCol("document")
val tokenizer = new Tokenizer()
.setInputCols(Array("document"))
.setOutputCol("token")
val sequenceClassifier = MedicalBertForSequenceClassification.pretrained("bert_sequence_classifier_covid_sentiment", "en", "clinical/models")
.setInputCols(Array("document","token"))
.setOutputCol("class")
val pipeline = new Pipeline().setStages(Array(documenter, tokenizer, sequenceClassifier))
val data = Seq(Array("British Department of Health confirms first two cases of in UK",
"so my trip to visit my australian exchange student just got canceled bc of coronavirus. im heartbroken :(",
"I wish everyone to be safe at home and stop pandemic"
)).toDS.toDF("text")
val result = pipeline.fit(data).transform(data)
import nlu
nlu.load("en.classify.bert_sequence.covid_sentiment").predict("""so my trip to visit my australian exchange student just got canceled bc of coronavirus. im heartbroken :(""")
Results
+---------------------------------------------------------------------------------------------------------+----------+
|text |result |
+---------------------------------------------------------------------------------------------------------+----------+
|British Department of Health confirms first two cases of in UK |[neutral] |
|so my trip to visit my australian exchange student just got canceled bc of coronavirus. im heartbroken :(|[negative]|
|I wish everyone to be safe at home and stop pandemic |[positive]|
+---------------------------------------------------------------------------------------------------------+----------+
Model Information
| Model Name: | bert_sequence_classifier_covid_sentiment |
| Compatibility: | Healthcare NLP 4.0.2+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [document, token] |
| Output Labels: | [class] |
| Language: | en |
| Size: | 406.5 MB |
| Case sensitive: | true |
| Max sentence length: | 128 |
References
Curated from several academic and in-house datasets.
Benchmarking
label precision recall f1-score support
negative 0.96 0.97 0.97 3284
positive 0.94 0.96 0.95 1207
neutral 0.96 0.94 0.95 3232
accuracy - - 0.96 7723
macro-avg 0.95 0.96 0.96 7723
weighted-avg 0.96 0.96 0.96 7723