Computing a group of related companies for financial information systems

• US 20070203720A1
• Filed: 02/24/2006
• Published: 08/30/2007
• Est. Priority Date: 02/24/2006
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Loiwal, Navneet, Singh, Amardeep, Narendran, Arungundram, Mohan, Bharath
• Granted Patent: US 7,873,595 B2
• Time in Patent Office: Days
• US Class Current: 705/7
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes

Abstract

Techniques are disclosed by which users looking for financial information about publicly traded or private companies may richly and interactively navigate both pricing and material news information about those companies. The techniques facilitate and encourage the user'"'"'s use and understanding of financial information presented. Related company information can also be provided to the user, where related companies are organized by hierarchal categories for a meaningful display.

Claims

1. A method for computing a group of related companies, comprising:

locating a co-occurrence of company names, including a target company and one or more other companies, using a first set of sources;

identifying relationships between the target company and each of the other companies;

verifying each identified relationship using sources not included in the first set of sources;

identifying hierarchical business categories to which the target company belongs; and

identifying one or more of the verified related companies associated with one or more of the identified business categories of the target company.

2. The method of claim 1 further comprising:

providing the identified one or more verified related companies in response to a user request for financial information about the target company.

3. The method of claim 1 wherein locating a co-occurrence of company names further comprises:

identifying all proper nouns in a given context; and

identifying company names included in the identified proper nouns.

4. The method of claim 1 wherein locating a co-occurrence of company names further comprises one or more of the following:

identifying known name variations of the target company;

identifying known aliases of the target company; and

disambiguating ambiguous terms based on co-occurrence context.

5. The method of claim 1 wherein identifying relationships between the target company and each of the other companies further comprises:

determining an overall score for a co-occurrence of the target company and another company, based on one or more relationship indicators; and

indicating a relationship between the target company and that other company if the overall score satisfies a relationship indicator threshold.

6. The method of claim 5 wherein determining the overall score includes one or more of the following:

determining if a co-occurrence context has many companies indicated at once;

determining where a co-occurrence is located in a document;

determining if a co-occurrence is located in a special context, including headlines or headings;

determining if a co-occurrence is reported by a reputable source;

determining a pagerank associated with a co-occurrence is located in a document; and

determining if a co-occurrence is associated with business relationship indicator words.

7. The method of claim 5 wherein verifying each identified relationship further comprises:

adjusting the overall score for an identified relationship, based on one or more supplemental relationship indicators; and

verifying the identified relationship if the adjusted overall score satisfies a verified relationship indicator threshold.

8. The method of claim 1 wherein verifying each identified relationship further comprises:

determining an overall score for an identified relationship, based on one or more relationship indicators; and

verifying the identified relationship if the overall score satisfies a relationship indicator threshold.

9. The method of claim 8 wherein determining the overall score includes one or more of the following:

determining if supplemental information sources indicate business, geography, or stock exchange in which co-occurring companies operate;

determining if supplemental information sources indicate at least one of competing products and services of occurring companies;

determining if supplemental information sources tend to treat co-occurring companies disparately;

determining if supplemental information sources that list businesses organized along hierarchical categories indicate co-occurring companies in one or more similar categories;

determining if supplemental information sources, including SEC filings, indicate co-occurring companies are direct competitors;

determining if supplemental information sources indicate an overlap of at least one of industries, sectors, and geographies in which co-occurring companies operate;

determining if supplemental information sources indicate a correlation between movements of stock prices of co-occurring companies;

determining if supplemental information sources indicate frequency of a co-occurrence exceeds a co-occurrence frequency threshold;

determining if supplemental information sources indicate overlap of co-occurring companies'"'"' products;

determining if supplemental information sources, including aggregated user navigation histories, indicate co-occurrences of companies; and

determining if supplemental information sources, including aggregated user portfolio data, indicate co-occurrences of companies.

10. The method of claim 1 wherein identifying one or more of the verified related companies comprises:

comparing one or more of the verified related companies to a list of companies associated with one or more of the identified hierarchical business categories to which the target company belongs, to identify matches.

11. The method of claim 10 wherein related companies provided in response to a user request include at least one of a high scoring verified related company and a verified related company that matched a company on the list.

12. A machine-readable medium encoded with instructions, that when executed by one or more processors, cause the processor to carry out a process for computing a group of related companies, the process comprising:

locating a co-occurrence of company names, including a target company and one or more other companies, using a first set of sources;

identifying relationships between the target company and each of the other companies;

verifying each identified relationship using sources not included in the first set of sources;

identifying hierarchical business categories to which the target company belongs; and

identifying one or more of the verified related companies associated with one or more of the identified business categories of the target company.

13. The machine-readable medium of claim 12, the process further comprising:

providing the identified one or more verified related companies in response to a user request for financial information about the target company.

14. The machine-readable medium of claim 12 wherein locating a co-occurrence of company names further comprises:

identifying all proper nouns in a given context; and

identifying company names included in the identified proper nouns.

15. The machine-readable medium of claim 12 wherein locating a co-occurrence of company names further comprises one or more of the following:

identifying known name variations of the target company;

identifying known aliases of the target company; and

disambiguating ambiguous terms based on co-occurrence context.

16. The machine-readable medium of claim 12 wherein identifying relationships between the target company and each of the other companies further comprises:

determining an overall score for a co-occurrence of the target company and another company, based on one or more relationship indicators; and

indicating a relationship between the target company and that other company if the overall score satisfies a relationship indicator threshold.

17. The machine-readable medium of claim 16 wherein determining the overall score includes one or more of the following:

determining if a co-occurrence context has many companies indicated at once;

determining where a co-occurrence is located in a document;

determining if a co-occurrence is located in a special context, including headlines or headings; and

determining if a co-occurrence is associated with business relationship indicator words.

18. The machine-readable medium of claim 16 wherein determining the overall score comprises:

determining if a co-occurrence is reported by a reputable source.

19. The machine-readable medium of claim 16 wherein determining the overall score comprises:

determining a pagerank associated with a co-occurrence is located in a document.

20. The machine-readable medium of claim 16 wherein verifying each identified relationship further comprises:

adjusting the overall score for an identified relationship, based on one or more supplemental relationship indicators; and

verifying the identified relationship if the adjusted overall score satisfies a verified relationship indicator threshold.

21. The machine-readable medium of claim 12 wherein verifying each identified relationship further comprises:

determining an overall score for an identified relationship, based on one or more relationship indicators; and

verifying the identified relationship if the overall score satisfies a relationship indicator threshold.

22. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate a business, geography, or stock exchange in which co-occurring companies operate.

23. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate at least one of competing products and services of occurring companies.

24. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources tend to treat co-occurring companies disparately.

25. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources that list businesses organized along hierarchical categories indicate co-occurring companies in one or more similar categories.

26. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources, including SEC filings, indicate co-occurring companies are direct competitors.

27. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate an overlap of at least one of industries, sectors, and geographies in which co-occurring companies operate.

28. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate a correlation between movements of stock prices of co-occurring companies.

29. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate frequency of a co-occurrence exceeds a co-occurrence frequency threshold.

30. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources indicate overlap of co-occurring companies'"'"' products.

31. The machine-readable medium of claim 21 wherein determining the overall score comprises:

determining if supplemental information sources, including at least one of aggregated user navigation histories and aggregated user portfolio data, indicate co-occurrences of companies.

32. The machine-readable medium of claim 12 wherein identifying one or more of the of verified related companies comprises:

comparing one or more of the verified related companies to a list of companies associated with one or more of the identified hierarchical business categories to which the target company belongs, to identify matches.

33. The machine-readable medium of claim 32 wherein related companies provided in response to a user request include at least one of a high scoring verified related company and a verified related company that matched a company on the list.

34. A system for computing a group of related companies, comprising:

a co-occurrence mining module for locating a co-occurrence of company names, including a target company and one or more other companies, using a first set of sources;

a relationship identifier module for identifying relationships between the target company and each of the other companies;

a relationship verifier module for verifying each identified relationship using sources not included in the first set of sources;

a hierarchal category identifier module for identifying hierarchical business categories to which the target company belongs; and

a hierarchal category placer module for identifying one or more of the verified related companies associated with one or more of the identified business categories of the target company.

35. The system of claim 34 wherein the relationship identifier module is further configured for determining an overall score for a co-occurrence of the target company and another company, based on one or more relationship indicators, and for indicating a relationship between the target company and that other company if the overall score satisfies a relationship indicator threshold.

36. The system of claim 35 wherein the relationship verifier module is further configured for adjusting the overall score for an identified relationship, based on one or more supplemental relationship indicators, and for verifying the identified relationship if the adjusted overall score satisfies a verified relationship indicator threshold.

37. The system of claim 34 wherein the relationship verifier module is further configured for determining an overall score for an identified relationship, based on one or more relationship indicators, and for verifying the identified relationship if the overall score satisfies a relationship indicator threshold.

38. The system of claim 34 wherein the hierarchal category placer module is further configured for comparing one or more of the verified related companies to a list of companies associated with one or more of the identified hierarchical business categories to which the target company belongs, to identify matches.

39. A system for computing a group of related companies, comprising:

means for locating a co-occurrence of company names, including a target company and one or more other companies, using a first set of sources;

means for identifying relationships between the target company and each of the other companies;

means for verifying each identified relationship using sources not included in the first set of sources;

means for identifying hierarchical business categories to which the target company belongs; and

means for identifying one or more of the verified related companies associated with one or more of the identified business categories of the target company.