Natural language processing for extracting conveyance graphs

• US 10,521,508 B2
• Filed: 12/30/2015
• Issued: 12/31/2019
• Est. Priority Date: 04/08/2014
• Status: Active Grant

• Current Assignee: Titleflow LLC
• Original Assignee: Titleflow LLC
• Inventors: Bateman, David T., Phillips, Aaron, Plagens, Andrew E., Drennan, J. Charles, Langdon, Wendell H.
• Primary Examiner(s): He, Jialong
• Application Number:
  US14/984,444
• Publication Number:
  US 20160117312A1
• Time in Patent Office: 1,462 Days
• Field of Search: 705 11, 705 4, 705 744, 705 727- 728, 705 1411, 705 30- 45, 705310-316, 705342
• CPC Class Codes: G06F 16/29 Geographical information databases
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 40/237 Lexical tools
• CPC Class Codes: G06F 40/279 Recognition of textual entities
• CPC Class Codes: G06Q 30/018 Certifying business or products
• CPC Class Codes: G06Q 30/0645 Rental transactions; Leasing transactions
• CPC Class Codes: G06Q 50/00 Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism healthcare informatics G16H
• CPC Class Codes: G06Q 50/02 Agriculture; Fishing; Forestry; Mining
• CPC Class Codes: G06Q 50/163 Real estate management
• CPC Class Codes: G06Q 50/165 Land development
• CPC Class Codes: G06V 30/413 Classification of content, e.g. text, photographs or tables
• CPC Class Codes: G06V 30/418 Document matching, e.g. of document images
• CPC Class Codes: G10L 21/00 Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility G10L19/00 takes precedence

Abstract

Provided is a process for extracting conveyance records from unstructured text documents, the process including: obtaining, with one or more processors, a plurality of documents describing, in unstructured form, one or more conveyances of interest in real property; determining, with one or more processors, for each of the documents, a respective jurisdiction; selecting, with one or more processors, from a plurality of language processing models for the English language, a respective language processing model for each of the documents based on the respective determined jurisdiction; extracting, with one or more processors, for each of the documents, a plurality of structured conveyance records from each of the plurality of documents by applying the language processing model selected for the respective document based on the jurisdiction associated with the document; and storing, with one or more processors, the extracted, structured conveyance record in memory.

Claims

1. A method of processing conveyance information obtained from repositories of unstructured data in county land offices, the method comprising:

obtaining and storing, with one or more processors, scanned documents having unstructured natural language text, each having at least some text describing one or more conveyances of interest in real property;

extracting, with one or more processors, with a natural language processing model, from the unstructured natural language text, parameters of a conveyance record, wherein;

the natural language processing model is responsive to features detected in unstructured natural language text based on hand-coded pattern-matching rules; and

the natural language processing model is trained by operations that comprise;

obtaining a training set of training documents and corresponding training records, the corresponding training records having respective conveyance record parameters appearing in the training documents, and

learning weights of the natural language processing model by training the natural language processing model to extract at least some conveyance record parameters based on the training set and features detected in the training documents based on the hand-coded pattern-matching rules;

obtaining and storing, with one or more processors, an ownership flow of an interest in real property described by at least some of the documents, the ownership flow including a plurality of conveyances in a conveyance graph, and the ownership flow including at least some of the extracted parameters;

receiving, from a client computing device, with one or more processors, a request to view the ownership flow, the request identifying the interest in real property;

applying, with one or more processors, ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules;

sending, to the client computing device, with one or more processors, instructions to display the ownership flow with an indication of a conveyance violating one or more of the rules;

receiving, from the client device, with one or more processors, a command to edit the ownership flow to comply with the violated rule;

in response to the command, with one or more processors, storing an updated ownership flow reflecting the edit;

after storing the updated ownership flow, receiving, with one or more processors, another request to view the ownership flow; and

sending, with one or more processors, instructions to display the updated ownership flow reflecting the edit.

2. The method of claim 1, wherein at least one hand-coded pattern-matching rule is a regular expression or nested conditional statement and at least some of the conveyance record parameters are extracted based on the hand-coded pattern-matching rules.

3. The method of claim 1, further comprising encoding the conveyance record in a standardized format, wherein the standardized format comprises key-value pairs that encode the extracted parameters of the conveyance record in JavaScript object notation (JSON) or extensible markup language (XML).

4. The method of claim 1, wherein each document is associated with a geolocation based on signals sensed by a mobile device used to scan the documents, the geolocation indicating a location of the documents at the time of scanning.

5. The method of claim 4, comprising verifying data manually entered structured data pertaining to a document for consistency with the geolocation associated with the document.

6. The method of claim 1, comprising:

receiving, from a client computing device, a request for owners of a type of interest in a geographic area; and

composing a response including a corresponding lists of owners; and

sending, to the client computing device, the corresponding list of owners.

7. The method of claim 6, wherein composing a response comprises identifying leaf nodes of an ownership graph.

8. The method of claim 1, comprising:

receiving, from a client computing device, a request for conveyance records listing a given party as a grantor or grantee;

composing a response including responsive conveyances; and

sending, to the client computing device, at least some of responsive conveyances.

9. The method of claim 1, comprising:

for at least a plurality of parties listed in the ownership flow, searching for other ownership flows listing the respective parties as a grantee or grantor; and

sending identifiers of responsive, other ownership flows to the client computing device.

10. The method of claim 9, wherein sending identifiers of the other ownership flows to the client computing device comprises:

sending links to images of documents effecting corresponding transfers of ownership.

21. The method of claim 1, wherein applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules comprises:

steps for interrogating the a directed graph to detect compliance with graph validity rules.

22. The method of claim 1, wherein applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules comprises:

determining that 100% of ownership of a type of interest is accounted for among leaf nodes of the conveyance graph.

23. The method of claim 1, wherein applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules comprises:

determining that 100% of ownership of each of a plurality of different types of interest are accounted for among leaf nodes of the conveyance graph.

24. The method of claim 1, wherein applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules comprises:

for each of a plurality of conveyances in the ownership flow, detecting compliance with a rule that a party can only convey that interest which they held at the time of the conveyance.

25. The method of claim 1, wherein applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules comprises:

detecting a severing conveyance in the conveyance graph in which mineral rights are severed from surface rights;

in response to detecting the severing conveyance, maintaining different inventories for mineral rights and surface rights and separately applying at least some ownership graph validity rules to the different inventories; and

associated a conveyance record that results in a violation of at least one ownership graph validity rule with an alarm designation identifying a type of the violation.

26. The method of claim 1, comprising:

steps for managing mineral rights.

11. A system, comprising:

one or more processors; and

memory storing instructions that when executed by at least some of the processors effectuate operations comprising;

obtaining and storing scanned documents having unstructured natural language text, each having at least some text describing one or more conveyances of interest in real property;

extracting, with one or more processors, with a natural language processing model, from the unstructured natural language text, parameters of a conveyance record, wherein;

the natural language processing model is responsive to features detected in unstructured natural language text based on hand-coded pattern-matching rules; and

the natural language processing model is trained by operations that comprise;

obtaining a training set of training documents and training records, the corresponding training records having respective conveyance record parameters appearing in the training documents, and

learning weights of the natural language processing model by training the natural language processing model to extract at least some conveyance record parameters based on the training set and features detected in the training documents based on the hand-coded pattern-matching rules;

obtaining and storing an ownership flow of an interest in real property described by at least some of the documents, the ownership flow including a plurality of conveyances in a conveyance graph;

receiving, from a client computing device, a request to view the ownership flow, the request identifying the interest in real property;

applying ownership graph rules to the ownership flow to identify conveyances in the ownership flow that violate one or more of the rules;

sending, to the client computing device, instructions to display the ownership flow with an indication of a conveyance violating one or more of the rules;

receiving, from the client device, a command to edit the ownership flow to comply with the violated rule;

in response to the command, storing an updated ownership flow reflecting the edit;

after storing the updated ownership flow, receiving another request to view the ownership flow; and

sending instructions to display the updated ownership flow reflecting the edit.

12. The system of claim 11, wherein at least one hand-coded pattern-matching rule is a regular expression or nested conditional statement and at least some of the conveyance record parameters are extracted based on the hand-coded pattern-matching rules.

13. The system of claim 11, further comprising encoding the conveyance record in a standardized format, wherein the standardized format comprises key-value pairs that encode the extracted parameters of the conveyance record in JavaScript object notation (JSON) or extensible markup language (XML).

14. The system of claim 11, wherein each document is associated with a geolocation based on signals sensed by a mobile device used to scan the documents, the geolocation indicating a location of the documents at the time of scanning.

15. The system of claim 14, comprising verifying data manually entered structured data pertaining to a document for consistency with the geolocation associated with the document.

16. The system of claim 11, the operations comprising:

receiving, from a client computing device, a request for owners of a type of interest in a geographic area; and

composing a response including a corresponding lists of owners; and

sending, to the client computing device, the corresponding list of owners.

17. The system of claim 16, wherein composing a response comprises identifying leaf nodes of an ownership graph.

18. The system of claim 11, the operations comprising:

receiving, from a client computing device, a request for conveyance records listing a given party as a grantor or grantee;

composing a response including responsive conveyances; and

sending, to the client computing device, at least some of responsive conveyances.

19. The system of claim 11, the operations comprising:

for at least a plurality of parties listed in the ownership flow, searching for other ownership flows listing the respective parties as a grantee or grantor; and

sending identifiers of responsive, other ownership flows to the client computing device.

20. The system of claim 19, wherein sending identifiers of the other ownership flows to the client computing device comprises:

sending links to images of documents effecting corresponding transfers of ownership.