Method and system for emulating web site traffic to identify web site usage patterns

• US 6,278,966 B1
• Filed: 06/18/1998
• Issued: 08/21/2001
• Est. Priority Date: 06/18/1998
• Status: Expired due to Fees

• Current Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Martin, David Charles, Howard, Steven Kenneth, Plutowski, Mark Earl Paul
• Primary Examiner(s): Teska, Kevin J.
• Assistant Examiner(s): Sergent, Douglas W.
• Application Number:
  US09/099,622
• Time in Patent Office: 1,160 Days
• Field of Search: 703/22-26, 709/224, 702/229, 714/27, 714/28
• US Class Current: 703/23
• CPC Class Codes: G06F 11/3414 Workload generation, e.g. scripts, playback
• CPC Class Codes: G06F 2201/875 Monitoring of systems including the internet

Abstract

A method and device to generate behavior for emulated visitors traversing an internet web site. The visitors may display behavior that is indistinguishable from those of actual users, a subset of the actual users, or the behavior may be purely hypothetical, such as when a visitor acts without evidence of having made an intentional choice. The invention tracks the actions of the visitors and develops reference distributions that may be compared to a site'"'"'s usage distributions as obtained from actual visitors to the site. The reference distributions are then used to implement statistical estimation methods that measure relative information content. The invention comprises a general implementation and a deterministic implementation. The general version may be applied to live production web sites, and the deterministic version is best suited to offline processing.

Claims

1. A method for emulating behavior of web site visitors for producing web site trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during a traversal of a web site and selecting a subset of a distribution to be emulated;

creating an emulated distribution including an entry page distribution, the emulated distribution emulating distribution and transition probabilities for selected actions of an emulated visitor;

specifying a maximum clickstream length;

storing the emulated distributions;

randomly selecting a number of visitors from the emulated distribution;

traversing a web site using the randomly selected emulated visitors; and

ending the emulation session.

13. The method of claim 1, the emulated traversal of a web site by a visitor comprising:

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site.

14. The method recited in claim 13, traversing a web site comprising:

generating a list of candidate links, where a candidate link is a link choice available on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

15. The method recited in claim 14, the method further comprising selecting only candidate links that are allowable links.

16. The method recited in claim 15, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

17. The method recited in claim 15, traversing a candidate link comprising:

enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

18. The method recited in claim 15, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

19. The method recited in claim 15, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

20. The method recited in claim 16, ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

21. The method recited in claim 17, ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

22. The method recited in claim 18, ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

23. The method recited in claim 19, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

2. A method for emulating behavior of web site visitors for producing web site trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during a traversal of a web site;

said emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site; and

randomly selecting a number of visitors from the emulated distribution;

traversing a web site using the randomly selected emulated visitors; and

ending the emulation session.

3. The method recited in claim 2, traversing the web site comprising:

generating a list of candidate links, a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

4. The method recited in claim 3, the method further comprising selecting only candidate links that are allowable links.

5. The method recited in claim 4, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

7. The method recited in claim 4, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

8. The method recited in claim 4, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

9. The method recited in claim 5, ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

11. The method recited in claim 7, ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

12. The method recited in claim 8, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

6. A method for emulating behavior of web site visitors for producing trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during a traversal of a web site;

randomly selecting a number of visitors from the emulated distribution;

traversing a web site using the randomly selected emulated visitors;

the emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page;

traversing the web site comprising generating a list of candidate links, a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list;

traversing the candidate link selecting only candidate links that are allowable links by enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links; and

ending the emulation session.

10. The method recited in claim 6 ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

24. A signal-bearing medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a method for emulating behavior of a web site visitor for producing web site trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during an emulated traversal of a web site and selecting a subset of a distribution to be emulated;

creating an emulated distribution including an entry page distribution, the emulated distribution emulating distribution and transition probabilities for selected actions of an emulated visitor;

specifying a maximum clickstream length; and

storing the emulated distribution;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

36. The medium of claim 24, the emulated traversal of a web site by a visitor comprising:

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site.

37. The medium recited in claim 36, traversing a web site comprising:

generating a list of candidate links, where a candidate link is a link choice available on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

38. The medium recited in claim 37, the method further comprising selecting only candidate links that are allowable links.

39. The medium recited in claim 38, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

40. The medium recited in claim 39, traversing a candidate link comprising:

enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

41. The medium recited in claim 39, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

42. The medium recited in claim 38, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

43. The medium recited in claim 39, ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

44. The medium recited in claim 40, ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

45. The medium recited in claim 43, ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

46. The medium recited in claim 44, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

25. A signal-bearing medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a method for emulating behavior of a web site visitor for producing web site trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during an emulated traversal of a web site;

said emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

26. The medium recited in claim 25, traversing the web site comprising:

generating a list of candidate links, a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

27. The medium recited in claim 26, the method further comprising selecting only candidate links that are allowable links.

28. The medium recited in claim 27, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

30. The medium recited in claim 27, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

31. The medium recited in claim 27, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

32. The medium recited in claim 28, ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

34. The medium recited in claim 30 ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

35. The medium recited in claim 31, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

29. A signal-bearing medium tangibly embodying a program of machine-readable instructions executable by a digital processing apparatus to perform a method for emulating behavior of a web site visitor for producing web site trend analysis data, the method comprising:

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by visitors during an emulated traversal of a web site;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors the emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page;

traversing the web site comprising;

generating a list of candidate links a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list;

traversing the candidate link comprising enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

selecting only candidate links that are allowable links; and

ending the emulation session.

33. The medium recited in claim 29, ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

47. A computer-driven system to emulated behavior of web site visitors for producing web site trend analysis data, the system comprising:

a storage;

a processor;

circuitry communicatively coupling the storage to the processor, the processor being capable of assisting in the emulation of web site visitor behavior by;

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by emulated visitors during an emulated traversal of a web site and selecting a subset of a distribution to be emulated;

creating an emulated distribution including an entry page distribution, the emulated distribution emulating distribution and transition probabilities for selected actions of an emulated visitor;

speciyfing a maximum clickstream length; and

storing the emulated distributions;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

59. The system of claim 47, the emulated traversal of a web site by a visitor comprising:

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site.

60. The system recited in claim 59, traversing a web site comprising:

generating a list of candidate links, where a candidate link is a link choice available on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

61. The system recited in claim 60, the method further comprising selecting only candidate links that are allowable links.

62. The system recited in claim 61, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

63. The system recited in claim 61, traversing a candidate link comprising:

enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

64. The system recited in claim 61, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

65. The system recited in claim 61, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

66. The system recited in claim 62, ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

67. The system recited in claim 63, ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

68. The system recited in claim 64, ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

69. The system recited in claim 65, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

48. A computer-driven system to emulated behavior of web site visitors for producing web site trend analysis data, the system comprising:

a storage;

a processor;

circuitry communicatively coupling the storage to the processor, the processor being capable of assisting in the emulation of web site visitor behavior by;

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by emulated visitors during an emulated traversal of a web site;

said emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

49. The system recited in claim 48, traversing the web site comprising:

generating a list of candidate links, a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list; and

traversing the candidate link.

50. The system recited in claim 49, the method further comprising selecting only candidate links that are allowable links.

51. The system recited in claim 50, traversing a candidate link comprising:

enabling link type preference distribution;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links.

53. The system recited in claim 50, traversing a candidate link comprising:

enabling global link preference distribution;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

54. The system recited in claim 50, traversing a candidate link comprising:

if link type preference distribution is enabled, then;

sorting candidate links by type;

weighing each candidate link using link preference distribution by link type, and ignoring any candidate link with a specified weight;

weighing candidate links by a uniform distribution where each candidate is equally as likely;

selecting allowable candidate links from the weighed candidate links; and

selecting at random an allowable candidate link from the allowable weighed candidate links;

if page preference distribution is enabled, then;

retrieving a link preference distribution for a current page is selected, then;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

if global link preference distribution is enabled, then;

sorting candidate links by position on a page;

weighing each candidate link using the global link preference distribution; and

selecting at random an allowable candidate link from the allowable weighted candidate links.

55. The system recited in claim 51 ending the emulation session comprising:

ending the emulation session if all allowable weighted link candidates have been traversed.

57. The system recited in claim 53, ending the emulation session comprising:

if an emulation session length has reached the global maximum, ending the session.

58. The system recited in claim 54, ending the emulation session comprising:

if a link preference distribution by link type is enabled, ending the emulation session if all allowable weighted link candidates have been traversed;

if the clickstream lifespan distribution is enabled, ending the emulation session;

if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution; and

if an emulation session length has reached the global maximum, ending the session.

52. A computer-driven system to emulated behavior of web site visitors for producing web site trend analysis data, the system comprising:

a storage;

a processor;

circuitry communicatively coupling the storage to the processor, the processor being capable of assisting in the emulation of web site visitor behavior by;

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by emulated visitors during an emulated traversal of a web site;

randomly selecting a number of emulated visitors from the emulated distribution by;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site comprising generating a list of candidate links, a candidate link being a link choice available to a visitor on a page of the web site;

selecting a candidate link from the list and selecting only candidate links that are allowable links; and

traversing the candidate link comprising enabling page preference distribution;

retrieving a link preference distribution for a current page;

weighing each candidate link using link preference distribution for the current page, and ignoring any candidate link with a specified weight; and

selecting at random an allowable candidate link from the allowable weighted candidate links;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

56. The system recited in claim 52, ending the emulation session comprising:

if the clickstream lifespan distribution is enabled, ending the emulation session;

otherwise, if a local clickstream lifespan distribution is enabled, choosing randomly whether or not to end the emulation session based upon a most relevant local distribution.

70. An apparatus for emulating behavior of web site visitors for producing web site trend analysis data, the apparatus comprising:

storage means for storing data;

a processing means for processing data, the processing means assisting in the emulation of web site visitor behavior by;

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by emulated visitors during an emulated traversal of a web site and selecting a subset of a distribution to be emulated;

creating an emulated distribution including an entry page distribution, the emulated distribution emulating distribution and transition probabilities for selected actions of an emulated visitor;

specifying a maximum clickstream length; and

storing the emulated distributions;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.

71. An apparatus for emulating behavior of web site visitors for producing web site trend analysis data, the apparatus comprising:

storage means for storing data;

a processing means for processing data, the processing means assisting in the emulation of web site visitor behavior by;

initializing an emulated distribution, the emulated distribution having data reflecting decisions made by emulated visitors during an emulated traversal of a web site, said emulated traversal of the web site by a visitor comprising;

selecting at random an entry page from an entry page distribution;

specifying a maximum clickstream length by randomly selecting a clickstream length from a clickstream lifespan distribution if the clickstream lifespan distribution is enabled;

entering the web site at the selected entry page; and

traversing the web site;

randomly selecting a number of emulated visitors from the emulated distribution;

traversing the web site using the randomly selected emulated visitors; and

ending the emulation session.