EXTRACTING, DERIVING, AND USING LEGAL MATTER SEMANTICS TO GENERATE E-DISCOVERY QUERIES IN AN E-DISCOVERY SYSTEM

• US 20200134757A1
• Filed: 10/30/2018
• Published: 04/30/2020
• Est. Priority Date: 10/30/2018
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Raphael, Roger C., Desai, Rajesh M., Islam, Nazrul, Hebbar, Satwik
• Granted Patent: US 11,132,755 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/2423 Interactive query statement specification based on a database schema
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/36 Creation of semantic tools, e.g. ontology or thesauri
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 40/242 Dictionaries
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06Q 50/18 Legal services

Abstract

Provided are techniques for extracting, deriving, and using legal matter semantics to generate e-discovery queries in an e-discovery system. A semantic knowledge graph is iteratively built by receiving meet and confer document instances, legal matter types, historical e-discovery queries for different legal matters, and legal semantic types extracted from the historical e-discovery queries. The legal semantic types are added to the semantic knowledge graph, and a list of terms that serve as a basis of an initial query are identified. An e-discovery query is generated for an e-discovery system. The e-discovery query is modified using the semantic knowledge graph and additional input by receiving a legal matter type and meet and confer information, obtaining the legal semantic types that are relevant to the legal matter type and the meet and confer information, and modifying the e-discovery query. The modified e-discovery query is provided. Then, the modified e-discovery query is executed.

Claims

1. A computer-implemented method for generating e-discovery queries for an e-discovery system comprising:

iteratively building a semantic knowledge graph during a build phase by,

receiving meet and confer document instances, legal matter types, historical e-discovery queries for different legal matters, and legal semantic types extracted from the historical e-discovery queries;

adding the legal semantic types to the semantic knowledge graph;

in response to adding the legal semantic types to the semantic knowledge graph,

identifying a list of terms that serve as a basis of an initial query; and

generating an e-discovery query for an e-discovery system by;

1) mapping a predicate clause of the initial query to a legal semantic type by;

identifying a metadata field and operator combination;

determining the legal semantic type from a glossary;

adding a first semantic type node to the semantic knowledge graph;

connecting the first semantic type node to a legal matter node with an edge weight; and

adding an expression term node for the metadata field with another edge weight; and

2) mapping a keyword of a text clause of the initial query to a legal semantic type by;

running a partial query for the keyword against a saved result set to generate a new result set; and

for each document in the new result set,

retrieving surrounding text around a location within the document where the keyword was found;

applying Named Entity Recognition (NER) to the keyword using surrounding text as context;

adding a second semantic type node for an entity type node;

connecting the second semantic type node to the legal matter node with an edge weight; and

adding an expression term node for the keyword with another edge weight; and

modifying the e-discovery query using the semantic knowledge graph and additional input during a query generation phase by;

receiving a legal matter type and meet and confer information;

issuing a series of questions to obtain the legal semantic types that are relevant to the legal matter type and the meet and confer information, wherein a priority of an order of questions is determined by weight associated with occurrence of those legal semantic types in the semantic knowledge graph;

modifying the e-discovery query based on the obtained legal semantic types to add one or more predicates that capture historical information based on user feedback responding to the series of questions;

providing the modified e-discovery query as a suggested query; and

in response to receiving selection of the modified e-discovery query, executing the modified e-discovery query.

2. The computer-implemented method of claim 1, wherein the meet and confer document instances are any of digital instances and physical instances.

3. The computer-implemented method of claim 1, further comprising:

providing a map for metadata fields and operator combinations.

4. The computer-implemented method of claim 1, further comprising:

providing a text index for documents.

5. The computer-implemented method of claim 1, further comprising:

receiving feedback on the modified e-discovery query to improve the semantic knowledge graph.

6. The computer-implemented method of claim 1, wherein a Software as a Service (SaaS) is configured to perform method operations.

7. A computer program product for generating e-discovery queries for an e-discovery system, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code executable by at least one processor to perform:

iteratively building a semantic knowledge graph during a build phase by,

receiving meet and confer document instances, legal matter types, historical e-discovery queries for different legal matters, and legal semantic types extracted from the historical e-discovery queries;

adding the legal semantic types to the semantic knowledge graph;

in response to adding the legal semantic types to the semantic knowledge graph,

identifying a list of terms that serve as a basis of an initial query; and

generating an e-discovery query to for e-discovery system by;

1) mapping a predicate clause of the initial query to a legal semantic type by;

identifying a metadata field and operator combination;

determining the legal semantic type from a glossary;

adding a first semantic type node to the semantic knowledge graph;

connecting the first semantic type node to a legal matter node with an edge weight; and

adding an expression term node for the metadata field with another edge weight; and

2) mapping a keyword of a text clause of the initial query to a legal semantic type by;

running a partial query for the keyword against a saved result set to generate a new result set; and

for each document in the new result set,

retrieving surrounding text around a location within the document where the keyword was found;

applying Named Entity Recognition (NER) to the keyword using the surrounding text as context;

adding a second semantic type node for an entity type node;

connecting the second semantic type node to the legal matter node with an edge weight; and

adding an expression term node for the keyword with another edge weight; and

modifying the e-discovery query using the semantic knowledge graph and additional input during a query generation phase by;

receiving a legal matter type and meet and confer information;

issuing a series of questions to obtain the legal semantic types that are relevant to the legal matter type and the meet and confer information, wherein a priority of an order of questions is determined by weight associated with occurrence of those legal semantic types in the semantic knowledge graph;

modifying the e-discovery query based on the obtained legal semantic types to add one or more predicates that capture historical information based on user feedback responding to the series of questions;

providing the modified e-discovery query as a suggested query; and

in response to receiving selection of the modified e-discovery query, executing the modified e-discovery query.

8. The computer program product of claim 7, wherein the meet and confer document instances are any of digital instances and physical instances.

9. The computer program product of claim 7, wherein the program code is executable by the at least one processor to perform:

providing a map for metadata fields and operator combinations.

10. The computer program product of claim 7, wherein the program code is executable by the at least one processor to perform:

providing a text index for documents.

11. The computer program product of claim 7, wherein the program code is executable by the at least one processor to perform:

receiving feedback on the modified e-discovery query to improve the semantic knowledge graph.

12. The computer program product of claim 7, wherein a Software as a Service (SaaS) is configured to perform computer program product operations.

13. A computer system for generating e-discovery queries for an e-discovery system, comprising:

one or more processors, one or more computer-readable memories and one or more computer-readable, tangible storage devices; and

program instructions, stored on at least one of the one or more computer-readable, tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, to perform operations comprising;

iteratively building a semantic knowledge graph during a build phase by,

receiving meet and confer document instances, legal matter types, historical e-discovery queries for different legal matters, and legal semantic types extracted from the historical e-discovery queries;

adding the legal semantic types to the semantic knowledge graph;

in response to adding the legal semantic types to the semantic knowledge graph, identifying a list of terms that serve as a basis of an initial query; and

generating an e-discovery query to for e-discovery system by;

1) mapping a predicate clause of the initial query to a legal semantic type by;

identifying a metadata field and operator combination;

determining the legal semantic type from a glossary;

adding a first semantic type node to the semantic knowledge graph;

connecting the first semantic type node to a legal matter node with an edge weight; and

adding an expression term node for the metadata field with another edge weight; and

2) mapping a keyword of a text clause of the initial query to a legal semantic type by;

running a partial query for the keyword against a saved result set to generate a new result set; and

for each document in the new result set,

retrieving surrounding text around a location within the document where the keyword was found;

applying Named Entity Recognition (NER) to the keyword using surrounding text as context;

adding a second semantic type node for an entity type node;

connecting the second semantic type node to the legal matter node with an edge weight; and

adding an expression term node for the keyword with another edge weight; and

modifying the e-discovery query using the semantic knowledge graph and additional input during a query generation phase by;

receiving a legal matter type and meet and confer information;

issuing a series of questions to obtain the legal semantic types that are relevant to the legal matter type and the meet and confer information, wherein a priority of an order of questions is determined by weight associated with occurrence of those legal semantic types in the semantic knowledge graph;

modifying the e-discovery query based on the obtained legal semantic types to add one or more predicates that capture historical information based on user feedback responding to the series of questions;

providing the modified e-discovery query as a suggested query; and

in response to receiving selection of the modified e-discovery query, executing the modified e-discovery query.

14. The computer system of claim 13, wherein the meet and confer document instances are any of digital instances and physical instances.

15. The computer system of claim 13, wherein the operations further comprise:

providing a map for metadata fields and operator combinations.

16. The computer system of claim 13, wherein the operations further comprise:

providing a text index for documents.

17. The computer system of claim 13, wherein the operations further comprise:

receiving feedback on the modified e-discovery query to improve the semantic knowledge graph.

18. The computer system of claim 13, wherein a Software as a Service (SaaS) is configured to perform computer system operations.