Description
This model is a Named Entity Recognition (NER) model focused on financial numeric items. It identifies 12 numeric financial entities from diverse 10-Q and 10-K reports. These entities are annotated using eXtensible Business Reporting Language (XBRL) tags. The annotation process primarily targets numerical tokens, and the context plays a crucial role in accurately assigning the appropriate entity type from the 139 most common financial entities available in the dataset.
This is a large (md) model, trained with 200K sentences.
Predicted Entities
PreferredStockSharesAuthorized, RelatedPartyTransactionExpensesFromTransactionsWithRelatedParty, PropertyPlantAndEquipmentUsefulLife, RestructuringCharges, PaymentsToAcquireBusinessesNetOfCashAcquired, ProceedsFromIssuanceOfCommonStock, PreferredStockDividendRatePercentage, RelatedPartyTransactionAmountsOfTransaction, RepaymentsOfDebt, RevenueFromContractWithCustomerExcludingAssessedTax, PublicUtilitiesRequestedRateIncreaseDecreaseAmount, RestructuringAndRelatedCostExpectedCost1
How to use
documentAssembler = nlp.DocumentAssembler() \
.setInputCol("text") \
.setOutputCol("document")
sentence = nlp.SentenceDetector() \
.setInputCols(["document"]) \
.setOutputCol("sentence")
tokenizer = nlp.Tokenizer()\
.setInputCols(["sentence"])\
.setOutputCol("token")\
.setContextChars(['.', ',', ';', ':', '!', '?', '*', '-', '(', ')', '”', '’', '$','€'])
embeddings = nlp.BertEmbeddings.pretrained("bert_embeddings_sec_bert_base","en") \
.setInputCols(["document", "token"]) \
.setOutputCol("embeddings")\
.setMaxSentenceLength(512)
nerTagger = finance.NerModel.pretrained('finner_10q_xbrl_md_subset9', 'en', 'finance/models')\
.setInputCols(["sentence", "token", "embeddings"])\
.setOutputCol("ner")
pipeline = nlp.Pipeline(stages=[documentAssembler,
sentence,
tokenizer,
embeddings,
nerTagger
])
text = "The Company recorded an expense of $ 407,094 in the second quarter of fiscal 2015 as a restructuring charge as an estimate for the difference between the rent that the Company pays its landlord and the rent received from the sub - tenant over the term of the sub - lease ."
df = spark.createDataFrame([[text]]).toDF("text")
fit = pipeline.fit(df)
result = fit.transform(df)
result_df = result.select(F.explode(F.arrays_zip(result.token.result,result.ner.result, result.ner.metadata)).alias("cols"))\
.select(F.expr("cols['0']").alias("token"),\
F.expr("cols['1']").alias("ner_label"),\
F.expr("cols['2']['confidence']").alias("confidence"))
result_df.show(50, truncate=100)
Results
+-------------+----------------------+----------+
|token |ner_label |confidence|
+-------------+----------------------+----------+
|The |O |1.0 |
|Company |O |1.0 |
|recorded |O |1.0 |
|an |O |1.0 |
|expense |O |1.0 |
|of |O |1.0 |
|$ |O |1.0 |
|407,094 |B-RestructuringCharges|0.997 |
|in |O |1.0 |
|the |O |1.0 |
|second |O |1.0 |
|quarter |O |1.0 |
|of |O |1.0 |
|fiscal |O |1.0 |
|2015 |O |1.0 |
|as |O |1.0 |
|a |O |1.0 |
|restructuring|O |1.0 |
|charge |O |1.0 |
|as |O |1.0 |
|an |O |1.0 |
|estimate |O |1.0 |
|for |O |1.0 |
|the |O |1.0 |
|difference |O |1.0 |
|between |O |1.0 |
|the |O |1.0 |
|rent |O |1.0 |
|that |O |1.0 |
|the |O |1.0 |
|Company |O |1.0 |
|pays |O |1.0 |
|its |O |1.0 |
|landlord |O |1.0 |
|and |O |1.0 |
|the |O |1.0 |
|rent |O |1.0 |
|received |O |1.0 |
|from |O |1.0 |
|the |O |1.0 |
|sub |O |1.0 |
|- |O |1.0 |
|tenant |O |1.0 |
|over |O |1.0 |
|the |O |1.0 |
|term |O |1.0 |
|of |O |1.0 |
|the |O |1.0 |
|sub |O |1.0 |
|- |O |1.0 |
|lease |O |1.0 |
|. |O |1.0 |
+-------------+----------------------+----------+
Model Information
| Model Name: | finner_10q_xbrl_md_subset9 |
| Compatibility: | Finance NLP 1.0.0+ |
| License: | Licensed |
| Edition: | Official |
| Input Labels: | [sentence, token, embeddings] |
| Output Labels: | [ner] |
| Language: | en |
| Size: | 16.4 MB |
References
An in-house modified version of https://huggingface.co/datasets/nlpaueb/finer-139, re-splited and filtered to focus on sentences with bigger density of tags.
Benchmarking
label precision recall f1-score support
B-PaymentsToAcquireBusinessesNetOfCashAcquired 0.9801 0.9610 0.9705 154
B-PreferredStockDividendRatePercentage 0.9822 1.0000 0.9910 166
B-PreferredStockSharesAuthorized 1.0000 1.0000 1.0000 113
B-ProceedsFromIssuanceOfCommonStock 0.9846 0.9014 0.9412 71
B-PropertyPlantAndEquipmentUsefulLife 0.9672 0.9743 0.9707 272
B-PublicUtilitiesRequestedRateIncreaseDecreaseAmount 1.0000 0.9894 0.9947 188
B-RelatedPartyTransactionAmountsOfTransaction 0.8750 0.3853 0.5350 218
B-RelatedPartyTransactionExpensesFromTransactionsWithRelatedParty 0.7215 0.9620 0.8245 447
B-RepaymentsOfDebt 0.9044 0.9762 0.9389 126
B-RestructuringAndRelatedCostExpectedCost1 0.8871 0.9483 0.9167 174
B-RestructuringCharges 0.9428 0.9450 0.9439 872
B-RevenueFromContractWithCustomerExcludingAssessedTax 0.9772 0.9062 0.9403 661
I-PreferredStockSharesAuthorized 1.0000 1.0000 1.0000 4
I-PropertyPlantAndEquipmentUsefulLife 1.0000 0.8171 0.8993 82
O 0.9989 0.9992 0.9990 77740
accuracy - - 0.9954 81288
macro-avg 0.9481 0.9177 0.9244 81288
weighted-avg 0.9957 0.9954 0.9952 81288