Method and system for displaying time-series data and correlated events derived from text mining

• US 20040027349A1
• Filed: 08/08/2002
• Published: 02/12/2004
• Est. Priority Date: 08/08/2002
• Status: Active Grant

• Current Assignee: Refinitiv US Organization, LLC (Refinitiv US Holdings, Inc.)
• Original Assignee: Reuters Limited (The Woodbridge Co. Ltd.)
• Inventors: Shemtov, Hadar, Feldman, Ronen, Landau, David, Lipshatz, Orly, Aumann, Yonatan
• Granted Patent: US 7,570,262 B2
• Time in Patent Office: Days
• US Class Current: 345/440
• CPC Class Codes: G06F 16/2474 Sequence data queries, e.g. querying versioned data
• CPC Class Codes: G06F 16/26 Visual data mining; Browsing structured data

Abstract

The present invention is directed to a method and system for correlating time-series data with events derived from text mining. The system is configured to receive a time period and a parameter concerning an entity, retrieve an event which is related to the entity and occurred within the time period from events which are previously extracted automatically from unstructured text, and display an indication of the event superimposed on a display representing the time series of the parameter for the time period.

Claims

1. A system for correlating data with events derived from text mining comprising:

an application server program recorded in an electronic memory and configured to;

(1) receive data indicating a time period and a parameter, the parameter concerning an entity, (2) retrieve an event previously extracted automatically from unstructured text, the event relating to the entity within the time period, and (3) display an indication of the event superimposed on a display representing the parameter during the time period.

2. The system of claim 1, wherein the application server program is further configured to retrieve data corresponding to the parameter within the time period.

3. The system of claim 2, further comprising a quantitative data repository for the application server program to retrieve data corresponding to the parameter.

4. The system of claim 3, wherein the quantitative data repository is a database or a connection to an external source.

5. The system of claim 1, wherein the application server program is further configured to receive an event type and the event belongs to the event type.

6. The system of claim 1, further comprising a knowledge base for storing a plurality of events previously extracted automatically from unstructured text.

7. The system of claim 1, further comprising a text analysis engine configured to automatically extract a plurality of events from unstructured text.

8. The system of claim 7, further comprising a text receiver configured to receive unstructured text.

9. The system of claim 7, wherein the unstructured text is a news article, a document, or a combination thereof.

10. The system of claim 7, wherein the text analysis engine is further configured to associated each of the plurality of the events with a time datum.

11. The system of claim 1, wherein the entity is a company which issues a stock and the parameter is the price of the stock.

12. The system of claim 11, wherein the event is an activity of the company, a competitor of the company, or a partner of the company.

13. The system of claim 11, wherein the event is a merger, an acquisition, a product introduction, a management change, or an earning warning.

14. The system of claim 1, wherein the application server program is further configured to correlate the parameter and the event.

15. The system of claim 1, wherein the indication of the event is an active icon and the application server program is further configured to:

receive a request indicating an activation of the icon;

retrieve information concerning the event upon receiving the request, the information derived from the unstructured text; and

display the information.

41. The system of claim 1, wherein the data is time-series data.

16. A method of correlating data with events derived from text mining, comprising:

receiving data indicating a time period and a parameter, the parameter concerning an entity;

retrieving an event previously extracted automatically from unstructured text, the activity concerning the entity within the time period; and

displaying an indication of the event superimposed on a display representing the parameter for the time period.

17. The method of claim 16, further comprising retrieving the parameter within the time period.

18. The method of claim 17, wherein the retrieving of the parameter is through a database or a connection to an external source.

19. The method of claim 16, further comprising receiving an event type, wherein the event belongs to the event type.

20. The method of claim 16, further comprising storing a plurality of events previously extracted automatically from unstructured text.

21. The method of claim 16, further comprising automatically extracting a plurality of events from unstructured text.

22. The method of claim 21, further comprising receiving unstructured text.

23. The method of claim 21, wherein the unstructured text is a news article, a document, or a combination thereof.

24. The method of claim 21, further comprising tagging each of the plurality of events with a time stamp.

25. The method of claim 16, wherein the entity is a company which issues a stock and the parameter is the price of the stock.

26. The method of claim 25, wherein the event is an activity of the company, a competitor of the company, or a partner of the company.

27. The method of claim 25, wherein the event is a merger, an acquisition, a product introduction, a management change, or an earning warning.

28. The method of claim 16, further comprising correlating the parameter and the event after retrieving the event and prior to displaying the indication of the event.

29. The method of claim 16, wherein the indication of the event is an active icon, the method further comprising:

receiving a request indicating an activation of the icon;

retrieving information concerning the event upon receiving the request, the information derived from the unstructured text; and

displaying the information.

42. The method of claim 16, wherein the data is time-series data.

30. A method of interacting with a computer for correlating data with events derived from text mining, comprising:

selecting a time period and a parameter, the parameter concerning an entity;

causing the computer to retrieve an event previously extracted automatically from unstructured text, the event concerning the entity within the time period; and

receiving an indication of the event superimposed on a display representing the parameter for the time period.

31. The method of claim 30, wherein the indication of the event is an active icon, the method further comprising:

sending a request to activate the icon;

causing the computer to retrieve information concerning the event, the information derived from the unstructured text; and

receiving a display of the information.

32. The method of claim 30, further comprising defining an event type, wherein the event belongs to the event type.

33. The method of claim 30, wherein the entity is a company which issues a stock and the parameter is the price of the stock.

34. The method of claim 33, wherein the event is an activity of the company, a competitor of the company, or a partner of the company.

35. The method of claim 33, wherein the event is a merger, an acquisition, a product introduction, a management change, or an earning warning.

43. The method of claim 30, wherein the data is time-series data.

36. A system for correlating data with events derived from text mining comprising:

an application server program recorded in an electronic memory and configured to;

(1) receive data indicating a time period and a parameter, the parameter concerning an entity, (2) retrieve an event previously extracted automatically from unstructured text, the event relating to the entity within the time period, and (3) create a data structure containing a temporal correlation between the parameter and the event for the time period.

44. The system of claim 36, wherein the data is time-series data.

37. A method of correlating data with events derived from text mining, comprising:

receiving data indicating a time period and a parameter, the parameter concerning an entity;

retrieving an event previously extracted automatically from unstructured text, the event concerning the entity within the time period; and

generating a data structure containing a temporal correlation between the parameter and the event within the time period.

45. The method of claim 37, wherein the data is time-series data.

38. A system for correlating a company'"'"'s stock price time-series with business events concerning the company, the business events derived from text mining, the system comprising:

a text receiver configured to receive unstructured text;

a text analysis engine configured to automatically extract business events from the unstructured text and tag each of the business events with a time stamp;

a knowledge base for storing the text and the business events;

a quantitative data repository for accessing stock information;

an interface configured to receive data indicating a time period and a stock;

an application server program recorded in an electronic memory and configured to (1) retrieve the price of the stock within the time period, (2) identify an activity from the business events, the activity concerning the company that issues the stock within the time period, (3) correlate the price and the activity, and (4) create a data structure containing the correlation between the price and the activity; and

a display for displaying the data structure in the form of a time series of the stock price superimposed with an active icon representing the activity.

39. A method of correlating a company'"'"'s stock price time-series with business events concerning the company, the business events derived from text mining, the method comprising:

receiving unstructured text;

extracting business events from the text;

tagging each of the business events with a time stamp;

storing the text and the business events;

receiving data indicating a time period and a stock;

retrieving the price of the stock within the time period;

identifying an activity that occurred within the time period from the business events, the activity concerning the company that issues the stock;

correlating the price and the activity;

generating a data structure containing the correlation between the price and the activity; and

displaying the data structure in the form of a time series of the stock price superimposed with an active icon representing the activity.

40. A method of interacting with a computer for correlating a company'"'"'s stock price time-series with business events concerning the company, the business events derived from text mining, the method comprising:

selecting a time period, an activity type, and a stock;

causing the computer to (1) search for the price of the stock within the time period, (2) identify an activity that occurred within the time period from a plurality of business events previously extracted automatically from unstructured text, the activity belonging to the activity type and concerning the company that issues the stock, and (3) create a data structure containing a correlation between the stock price and the activity;

receiving a display of the data structure in the form of a time series of the stock price superimposed with an active icon representing the activity;

sending a request to activate the icon; and

receiving a display of information concerning the activity, the information derived from the unstructured text.