Cluster-based visualization of user traffic on an internet site

• US 7,454,705 B2
• Filed: 04/22/2004
• Issued: 11/18/2008
• Est. Priority Date: 03/02/2000
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Meek, Christopher A., Heckerman, David E., Cadez, Igor, White, Steven J.
• Primary Examiner(s): Chow; Dennis-Doon
• Assistant Examiner(s): Ulrich; Nicholas S
• Application Number:
  US10/829,548
• Publication Number:
  US 20040196311A1
• Time in Patent Office: 1,671 Days
• Field of Search: 709/224, 707/7, 715/700
• US Class Current: 715/700
• CPC Class Codes: H04L 41/0893 Assignment of logical groups to network elements
• CPC Class Codes: H04L 41/22 comprising specially adapted graphical user interfaces [GUI]
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

Visualizing Internet web traffic is disclosed. In one embodiment, a number of windows are displayed, corresponding to a number of clusters into which users have been partitioned based on similar web browsing behavior. The windows are ordered from the cluster having the greatest number of users to the cluster having the least number of users. Each window has one or more rows, where each row corresponds to a user within the cluster. Each row has an ordered number of visible units, such as blocks, where each block corresponds to a web page visited by the user. The blocks can be color coded by the type of web page they represent. In one embodiment, the corresponding cluster models for the clusters are alternatively displayed in the windows.

Claims

1. A system that visualizes web site activity traffic, comprising:

program modules embodied on a computer-readable storage medium, the program modules including the following components;

a monitoring component that obtains information related to browsing activity of users visiting a web site;

a component that analyzes the information and parses the users into one or more user groups based on the analyzed information; and

a visualization component that graphically presents user browsing information in one or more windows within a display screen, the one or more windows corresponding to the one or more user groups, respectively, and the browsing information being displayed within a window corresponding to the user group associated with the window, the browsing information within the window being delineated by user into one or more rows such that each row corresponds to an individual user, the one or more rows ordered according to a probability of a behavior based on a location of a user within a group.

2. The system of claim 1, the groups being defined via a non-restrictive and/or a non-limiting set of similar items that are associated with one another based on one or more common or similar characteristics.

3. The system of claim 1, the one or more rows comprising one or more units that store information associated with web pages visited by the users.

4. The system of claim 3, the respective units within a row being color coded according to a type of web page represented.

5. The system of claim 4, the type representing one or more of local news, global news, financial news, and entertainment.

6. The system of claim 3, the respective units being associated with intensity levels that indicate a frequency with which a user visits a type of web page.

7. The system of claim 6, the frequency comprising a probability of visiting a type of web page.

8. The system of claim 6, the intensity levels ranging from low to high, a low intensity indicating a low probability of visiting a web page and a high intensity indicating a high probability of visiting the web page.

9. The system of claim 1, the respective rows being associated with an intensity level that indicates a conditional transition probability of a zero-order or a first-order Markov model.

10. The system of claim 1, the respective rows being associated with labels that correspond to a type of web page represented by the row.

11. The system of claim 1, the windows being sorted by a number of users within the group associated with the windows.

12. The system of claim 1, the component utilizing an expectation-maximization (EM) algorithm to facilitate generating the groups.

13. The system of claim 1, the respective windows being configured to change size to display more rows within a visible region of the windows.

14. The system of claim 1, the respective windows including a scroll bar that is utilized to navigate through the rows data displayed within the windows.

15. A method that displays web traffic, comprising:

receiving web site user clusters that were generated through one of zero-order and first-order Markov models, respective clusters including information related to one or more web pages accessed by one or more users who display similar web browsing characteristics;

creating individual graphical user interfaces for each cluster;

visualizing the cluster information within one or more rows of an associated graphical user interface; and

sorting the one or more rows according to a probability of a behavior based on a location of a user within a cluster.

16. The method of claim 15, further comprising defining the clusters with a non-restrictive and a non-limiting group of associated items.

17. The method of claim 15, further comprising delineating the one or more rows into one or more units that respectively store the web page information.

18. The method of claim 17, further comprising employing a color code with the units to differentiate respective units according to a type of web page represented.

19. A method for displaying web site user activity according to web site behavior, comprising:

utilizing an expectation-maximization algorithm to cluster web site users by browsing behavior;

displaying user web site activity by visualizing information related to web pages accessed by users with similar behavior in fields displayed within a window; and

employing different intensity levels with respective fields to indicate a probability of visiting a type of web page.

20. The method of claim 19, further comprising labeling respective fields by a type of web page represented.

21. The method of claim 20, the type representing one or more of local news, global news, financial news, and entertainment.

22. The system of claim 19, further comprising ordering the windows by a number of users within a cluster.

23. A data packet embodied on a computer-readable storage medium that when transmitted between two or more computer components that facilitates visualizing web site activity, comprising:

a plurality of clusters that respectively include users with similar browsing behavior and web pages visited by the users, wherein the information within the plurality of clusters is displayed in windows, based on respective clusters, of a display and the web page information is partitioned into rows within the windows such that each row is associated with an individual user, the one or more rows ordered according to a probability of a behavior based on a location of a user in a cluster.

24. A computer readable medium storing computer executable components that facilitates visualizing web site activity, comprising:

a component that partitions web site users according to similar web site navigation;

a component that generates a window for each partition within a display;

a component that displays information relating to web pages visited by the users in associated windows; and

a component that sorts the web pages information within fields of a plurality of rows that respectively correspond to individual users, the fields displayed employing different intensity levels to indicate a probability of a user visiting a type of web page, the intensity levels ranging from low to high, a low intensity indicating a low probability of visiting the web page, and a high intensity indicating a high probability of visiting the web page.

25. A system that facilitates visualization web site activity, the system comprising program modules embodied on a computer-readable storage medium that facilitate:

means for clustering web site activity information according to web site user behavior;

means for displaying the clustered web site activity information in respective windows;

means for presenting the web site activity information within respective windows based on similar user browsing behavior; and

means for delineating the web site activity information into one or more rows within the respective windows, each row represents an individual user of the web site, the respective rows associated with an intensity level indicating a conditional transition probability of a zero-order or a first-order Markov model.