Description
This is a Financial Entity Resolver model, trained to obtain the ticker from a Company Name, registered in NASDAQ. You can use this model after extracting a company name using any NER, and you will obtain its ticker.
After this, you can use finmapper_nasdaq_data_ticker to augment and obtain more information about a company using NASDAQ datasource, including Official Company Name, Sector, Location, Currency, etc.
Predicted Entities
How to use
test = ["FIDUS INVESTMENT corp","ASPECT DEVELOPMENT Inc","CFSB BANCORP","DALEEN TECHNOLOGIES","GLEASON Corporation"]
testdf = pandas.DataFrame(test, columns=['text'])
testsdf = spark.createDataFrame(testdf).toDF('text')
documentAssembler = nlp.DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("sentence")
use = nlp.UniversalSentenceEncoder.pretrained("tfhub_use_lg", "en")\
.setInputCols(["sentence"])\
.setOutputCol("embeddings")
use_er_model = finance.SentenceEntityResolverModel.pretrained('finel_nasdaq_data_ticker', 'en', 'finance/models')\
.setInputCols("embeddings")\
.setOutputCol('normalized')\
prediction_Model = nlp.Pipeline(stages=[documentAssembler, use, use_er_model])
test_pred = prediction_Model.transform(testsdf).cache()
Results
+----------------------+-------+
|text |result |
+----------------------+-------+
|FIDUS INVESTMENT corp |[FDUS] |
|ASPECT DEVELOPMENT Inc|[ASDV] |
|CFSB BANCORP |[CFSB] |
|DALEEN TECHNOLOGIES |[DALN1]|
|GLEASON Corporation |[GLE1] |
+----------------------+-------+
Model Information
| Model Name: | finel_nasdaq_data_ticker |
| Compatibility: | Finance NLP 1.0.0+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [embeddings] |
| Output Labels: | [normalized] |
| Language: | en |
| Size: | 69.8 MB |
| Case sensitive: | false |
References
NASDAQ Database