Description
IMPORTANT: Don’t run this model on the whole legal agreement. Instead:
- Split by paragraphs. You can use notebook 1 in Finance or Legal as inspiration;
- Use the
legclf_introduction_clauseText Classifier to select only these paragraphs;
This is a Legal Relation Extraction model, which can be used after the NER Model for extracting Parties, Document Types, Effective Dates and Aliases, called legner_contract_doc_parties.
As an output, you will get the relations linking the different concepts together, if such relation exists. The list of relations is:
- dated_as: A Document has an Effective Date
- has_alias: The Alias of a Party all along the document
- has_collective_alias: An Alias hold by several parties at the same time
- signed_by: Between a Party and the document they signed
This model is a sm model without meaningful directions in the relations (the model was not trained to understand if the direction of the relation is from left to right or right to left). There are bigger models in Models Hub trained also with directed relationships.
Predicted Entities
dated_as, has_alias, has_collective_alias, signed_by
How to use
documentAssembler = nlp.DocumentAssembler()\
.setInputCol("text")\
.setOutputCol("document")
tokenizer = nlp.Tokenizer()\
.setInputCols("document")\
.setOutputCol("token")
embeddings = nlp.BertEmbeddings.pretrained("bert_base_uncased_legal", "en") \
.setInputCols("document", "token") \
.setOutputCol("embeddings")
ner_model = legal.NerModel.pretrained('legner_contract_doc_parties', 'en', 'legal/models')\
.setInputCols(["document", "token", "embeddings"])\
.setOutputCol("ner")
ner_converter = nlp.NerConverter()\
.setInputCols(["document","token","ner"])\
.setOutputCol("ner_chunk")
reDL = legal.RelationExtractionDLModel().pretrained('legre_contract_doc_parties', 'en', 'legal/models')\
.setPredictionThreshold(0.5)\
.setInputCols(["ner_chunk", "document"])\
.setOutputCol("relations")
nlpPipeline = nlp.Pipeline(stages=[
documentAssembler,
tokenizer,
embeddings,
ner_model,
ner_converter,
reDL
])
text='''
This INTELLECTUAL PROPERTY AGREEMENT (this "Agreement"), dated as of December 31, 2018 (the "Effective Date") is entered into by and between Armstrong Flooring, Inc., a Delaware corporation ("Seller") and AFI Licensing LLC, a Delaware limited liability company ("Licensing" and together with Seller, "Arizona") and AHF Holding, Inc. (formerly known as Tarzan HoldCo, Inc.), a Delaware corporation ("Buyer") and Armstrong Hardwood Flooring Company, a Tennessee corporation (the "Company" and together with Buyer the "Buyer Entities") (each of Arizona on the one hand and the Buyer Entities on the other hand, a "Party" and collectively, the "Parties").
'''
data = spark.createDataFrame([[text]]).toDF("text")
model = nlpPipeline.fit(data)
Results
relation entity1 entity1_begin entity1_end chunk1 entity2 entity2_begin entity2_end chunk2 confidence
dated_as DOC 6 36 INTELLECTUAL PROPERTY AGREEMENT EFFDATE 70 86 December 31, 2018 0.9933402
signed_by DOC 6 36 INTELLECTUAL PROPERTY AGREEMENT PARTY 142 164 Armstrong Flooring, Inc 0.6235637
signed_by DOC 6 36 INTELLECTUAL PROPERTY AGREEMENT PARTY 316 331 AHF Holding, Inc 0.5001139
has_alias PARTY 142 164 Armstrong Flooring, Inc ALIAS 193 198 Seller 0.93385726
has_alias PARTY 206 222 AFI Licensing LLC ALIAS 264 272 Licensing 0.9859913
has_collective_alias ALIAS 293 298 Seller ALIAS 302 308 Arizona 0.82137156
has_alias PARTY 316 331 AHF Holding, Inc ALIAS 400 404 Buyer 0.8178999
has_alias PARTY 412 446 Armstrong Hardwood Flooring Company ALIAS 479 485 Company 0.9557921
has_alias PARTY 412 446 Armstrong Hardwood Flooring Company ALIAS 575 579 Buyer 0.6778585
has_alias PARTY 412 446 Armstrong Hardwood Flooring Company ALIAS 612 616 Party 0.6778583
has_alias PARTY 412 446 Armstrong Hardwood Flooring Company ALIAS 642 648 Parties 0.6778585
has_collective_alias ALIAS 506 510 Buyer ALIAS 517 530 Buyer Entities 0.69863707
has_collective_alias ALIAS 517 530 Buyer Entities ALIAS 575 579 Buyer 0.55453944
has_collective_alias ALIAS 517 530 Buyer Entities ALIAS 612 616 Party 0.55453944
has_collective_alias ALIAS 517 530 Buyer Entities ALIAS 642 648 Parties 0.55453944
Model Information
| Model Name: | legre_contract_doc_parties |
| Type: | legal |
| Compatibility: | Legal NLP 1.0.0+ |
| License: | Licensed |
| Edition: | Official |
| Language: | en |
| Size: | 409.9 MB |
References
Manual annotations on CUAD dataset
Benchmarking
label Recall Precision F1 Support
dated_as 0.962 0.962 0.962 26
has_alias 0.936 0.946 0.941 94
has_collective_alias 1.000 1.000 1.000 7
no_rel 0.982 0.980 0.981 497
signed_by 0.961 0.961 0.961 76
Avg. 0.968 0.970 0.969 -
Weighted-Avg. 0.973 0.973 0.973 -