System and method for automatic identification of legal entities
Systems, methods, and computer readable media for identifying entities as legal entities are provided. These techniques may include accessing a corpus of documents and applying a persona prediction machine learning algorithm to classify entities associated with the corpus of documents. The persona prediction machine learning algorithm may include two layers. A first layer includes applying a signature block classifier that analyzes signature blocks of the entities. A second layer includes applying an entity classifier that analyzes a plurality of documents and/or network graphs associated with the entities. An entity database is updated to indicate the output of the persona prediction machine learning algorithm based on the signature block classifier and/or the entity classifier.
1. A computer-implemented method for identifying legal entities, the method comprising:
accessing, by one or more processors, a corpus of documents;
accessing, by the one or more processors, an entity database that includes a plurality of records respectively corresponding to a plurality of entities indicated by documents in the corpus of documents;
executing, by the one or more processors, a persona prediction machine learning algorithm on entities included in the plurality of entities, the persona prediction machine learning algorithm including:
a first layer configured to apply a signature block classifier that analyzes signature blocks corresponding to a particular entity to determine whether the particular entity is a legal entity; and
a second layer configured to apply an entity classifier that analyzes a set of documents associated with the entity and a network graph for the entity to determine whether the particular entity is a legal entity, wherein executing the persona prediction machine learning algorithm comprises:
determining, by the one or more processors, that the signature block classifier does not meet a threshold confidence in determining whether or not the particular entity is a legal entity, and
in response to the determination, applying, by the one or more processors, the second layer to the particular entity; and
updating, by one or more processors, the entity database to indicate whether the entities are legal entities based on outputs of the persona prediction machine learning algorithm.
2. The computer-implemented method of claim 1 , further comprising:
analyzing, by the one or more processors, documents of the corpus of documents associated with the particular entity to identify a signature block corresponding to the particular entity; and
updating, by the one or more processors, the record in the entity database corresponding to the particular entity to indicate the signature block for the particular entity.
3. The computer-implemented method of claim 1 , wherein performing the first layer of the persona prediction machine learning algorithm comprises:
applying, by the one or more processors, a parser to parse the signature block into one or more features of the signature block; and
applying, by the one or more processors, the signature block classifier to classify the particular entity as a legal entity based upon the one or more features of the signature block.
4. The computer-implemented method of claim 3 , wherein the features of the signature block include an indication of one or more of a title, a credential, an email address domain, or a confidentiality notice.
5. The computer-implemented method of claim 3 , wherein the parser is trained via a support vector machine (SVM) model and the signature block classifier is trained via a convolutional neural network (CNN) model.
6. The computer-implemented method of claim 1 , wherein applying the entity classifier comprises:
generating, by the one or more processors, a feature corresponding to a percentage of first-degree communications for the particular persona with other personas that are legal entities.
7. The computer-implemented method of claim 1 , wherein applying the entity classifier comprises:
identifying, by the one or more processors, one or more topics corresponding to documents in the corpus of documents associated with the particular entity; and
based on the identified topics, generating, by the one or more processors, one or more features based upon the one or more identified topics.
8. The computer-implemented method of claim 1 , wherein the entity classifier is trained based upon a logistic regression model.
9. The computer-implemented method of claim 1 , further comprising:
presenting, by the one or more processors, a plurality of outputs of the persona prediction machine learning algorithm to a user for manual review;
detecting, by the one or more processors, an indication of whether or not the persona prediction machine learning algorithm correctly determined whether entities are legal entities;
based on the detected indications, re-training, by the one or more processors, at least one of the signature block classifier or the entity classifier.
10. A system for automatically identifying legal entities, the system comprising:
one or more processors;
a communication interface communicatively coupled to a document storage system storing a corpus of documents;
an entity database configured to store a plurality of records respectively corresponding to entities indicated by the corpus of documents; and
one or more memories storing non-transitory, computer-readable instructions that, when executed by the one or more processors, cause the system to:
access, via the communication interface, the corpus of documents;
execute a persona prediction machine learning algorithm on entities associated with records in the entity database, the persona prediction machine learning algorithm including:
a first layer configured to apply a signature block classifier that analyzes signature blocks corresponding to a particular entity to determine whether the particular entity is a legal entity; and
a second layer configured to apply an entity classifier that analyzes a set of documents associated with the entity and a network graph for the entity to determine whether the particular entity is a legal entity, wherein the one or more processors executing the persona prediction machine learning algorithm further comprises causing the one or more processors to:
determine that the signature block classifier does not meet a threshold confidence in determining whether or not the particular entity is a legal entity, and
in response to the determination, applying the second layer to the particular entity; and
update the entity database to indicate whether the entities are legal entities based on outputs of the persona prediction machine learning algorithm.
11. The system of claim 10 , wherein the instructions, when executed, cause the system to:
analyze documents of the corpus of documents associated with the particular entity to identify a signature block corresponding to the particular entity; and
update the record in the entity database corresponding to the entity to indicate the signature block for the particular entity.
12. The system of claim 10 , wherein the first layer comprises:
a parser configured to parse the signature block into one or more features of the signature block; and
the signature block classifier, wherein the signature block classifier is configured to classify the particular entity as a legal entity based upon the one or more features of the signature block.
13. The system of claim 12 , wherein the parser is trained via a support vector machine (SVM) model and the signature block classifier is trained via a convolutional neural network (CNN) model.
14. The system of claim 10 , wherein to apply the entity classifier, the instructions, when executed, cause the system to:
generate a feature corresponding to a percentage of first-degree communications for the particular persona with other personas that are legal entities.
15. The system of claim 10 , wherein to apply the entity classifier, the instructions, when executed, cause the system to:
identify one or more topics corresponding to documents in the corpus of documents associated with the particular entity; and
based on the identified topics, generate one or more features based upon the one or more identified topics.
16. The system of claim 10 , wherein the entity classifier is trained based upon a logistic regression model.
17. The system of claim 10 , wherein the instructions, when executed, cause the system to:
present a plurality of outputs of the persona prediction machine learning algorithm to a user for manual review;
detect an indication of whether or not the machine learning algorithm correctly determined whether entities are legal entities;
based on the detected indications, re-train at least one of the signature block classifier or the entity classifier.
18. A non-transitory computer-readable storage medium storing processor-executable instructions, that when executed cause one or more processors to:
access a corpus of documents;
access an entity database that includes a plurality of records respectively corresponding a plurality of entities indicated by documents in the corpus of documents;
execute a persona prediction machine learning algorithm on entities included in the plurality of entities, the persona prediction machine learning algorithm including:
a first layer configured to apply a signature block classifier that analyzes signature blocks corresponding to a particular entity to determine whether the particular entity is a legal entity; and
a second layer configured to apply an entity classifier that analyzes a set of documents associated with the entity and a network graph for the entity to determine whether the particular entity is a legal entity, wherein the one or more processors executing the persona prediction machine learning algorithm further comprises causing the one or more processors to:
determine that the signature block classifier does not meet a threshold confidence in determining whether or not the particular entity is a legal entity, and
in response to the determination, applying the second layer to the particular entity; and
update the entity database to indicate whether the entities are legal entities based on outputs of the persona prediction machine learning algorithm.