FINANCIAL EVENT AND RELATIONSHIP EXTRACTION

• US 20090327115A1
• Filed: 01/30/2009
• Published: 12/31/2009
• Est. Priority Date: 01/30/2008
• Status: Active Grant

• Current Assignee: Thomson Reuters Enterprise Centre GmbH (The Woodbridge Co. Ltd.)
• Original Assignee: Thomson Reuters Global Resources
• Inventors: Shaw, James, Schilder, Frank
• Granted Patent: US 10,049,100 B2
• Time in Patent Office: Days
• US Class Current: 705/35
• 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 40/295 Named entity recognition
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03

Abstract

For automated text processing, the inventors devised, among other things, an exemplary system that automatically extracts financial events from various unstructured text based sources, such as press releases and news articles. Extracted events, such as mergers & acquisitions, earnings guidance reports, and actual earnings announcements, are represented as structured data records which can be linked, searched, and displayed and used as a basis for controlling accessing to the source documents and other related financial documents for named entities.

Claims

1. A computer system having at least one processor and at least one memory, the system comprising:

means for automatically tagging entity names, monetary expressions, and temporal expressions within a text segment;

means for determining whether the automatically tagged text segment describes a financial event; and

means, responsive to an affirmative determination of a financial event in the tagged text segment, for defining in memory a data record including data derived from the tagged text segment.

2. The system of claim 1, wherein the text segment is a grammatical sentence.

3. The system of claim 1, wherein the data record includes:

a company field including text identifying a named entity tagged in the text segment;

a company ID field including a alphanumeric code identifying the named entity; and

a time period field including an alphanumeric code identifying a financial reporting period.

4. The system of claim 1, wherein the data record includes a field indicating whether a monetary expression tagged in the text segment is trending up or down.

5. The system of claim 1, wherein the data record includes a field indicating that a monetary expression tagged in the text segment is a measure of earnings per share.

6. The system of claim 1, wherein the means for automatically tagging entity names, monetary expressions, and temporal expressions within a text segment includes:

first means for tagging and resolving entity names;

second means for tagging monetary expression; and

third means for tagging temporal expressions.

7. The system of claim 1, wherein the means for determining whether the automatically tagged text segment describes a financial event includes a mergers &

acquisition (M&

A) classifier for determining whether text segments describe an M&

A event.

8. The system of claim 1, wherein the means for determining whether the automatically tagged text segment describes a financial event includes a mergers &

acquisition (M&

A) classifier for determining whether or not text segments describe an M&

A event.

9. The system of claim 8, wherein the M&

A classifier is machine-learning based classifier.

10. The system of claim 1, wherein the means for determining whether the automatically tagged text segment describes a financial event includes a an earnings event classifier for determining whether or not the text segment describes an earnings event.

11. The system of claim 1, wherein the means for determining whether the automatically tagged text segment describes a financial event includes a guidance event classifier for determining whether or not the text segment describes a financial guidance event.

12. A computer implemented method comprising:

automatically tagging entity names, monetary expressions, and temporal expressions within a text segment;

determining whether the automatically tagged text segment describes a financial event; and

in response to an affirmative determination of a financial event in the tagged text segment, defining in memory a data record including data derived from the tagged text segment.

13. The method of claim 12, further comprising displaying on a display device at least a portion of the data record in association with a user selectable command feature of a graphical user interface for causing retrieval of a document including the text segment.

14. The method of claim 12, wherein the text segment is a grammatical sentence.

15. The method of claim 12, wherein the data record includes:

a company field including text identifying a named entity tagged in the text segment;

a company ID field including a alphanumeric code identifying the named entity; and

a time period field including an alphanumeric code identifying a financial reporting period.

16. The method of claim 12, wherein the data record includes a field indicating whether a monetary expression tagged in the text segment is trending up or down;

automatically tagging entity names within a text segment as being one of a person, company, and location; and

automatically associating one or more of the tagged entity names with an entry in a data set of named entities.

17. A computer-implemented method of identifying and extracting financial information from tables in documents, the method comprising:

identifying within a document a table that contains at least one predetermined desired relation;

partitioning the identified table into one or more labels and one or more values, with one or more of the labels identified as a column label and one or more identified as a row label;

associating each value of the one more values partitioned from the table with one or more of the labels, with the abstract table including a list of attribute-value pairs; and

populating a database of financial information based on the list of attribute-value pairs.

18. The method of claim 17, wherein identifying in the document a table that contains at least one predetermined desired relationship includes using a support vector machine.

19. The method of claim 17, wherein the document is a Securities and Exchange Commission filing.