System and methods for automatically creating lists

• US 7,350,187 B1
• Filed: 04/30/2003
• Issued: 03/25/2008
• Est. Priority Date: 04/30/2003
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Dean, Jeff, Tong, Simon
• Primary Examiner(s): Khatri; Anil
• Application Number:
  US10/425,819
• Time in Patent Office: 1,791 Days
• Field of Search: 717104-113, 717/120, 717/123, 717/125, 717/144, 707/100, 707/203, 707 2- 4, 709/203
• US Class Current: 717/104
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/274 Converting codes to words; Guess-ahead of partial word inputs

Abstract

A system automatically creates a list from items in existing lists. The system receives one or more example items corresponding to the list and assigns weights to the items in the existing lists based on the one or more example items. The system then forms the list based on the items and the weights assigned to the items.

Claims

1. A method, comprising:

determining an on-topic model;

identifying a plurality of lists within one or more documents;

determining probabilities of the plurality of lists being generated using the on-topic model;

forming a list from items in the plurality of lists based on the determined probabilities; and

presenting the list to a client.

2. The method of claim 1, further comprising:

receiving one or more example items; and

wherein the determining an on-topic model includes;

determining a probabilistic model based on the one or more example items.

3. The method of claim 2, wherein the determining a probabilistic model includes:

assigning higher probabilities to the one or more example items than to other items.

4. The method of claim 1, wherein identifying the plurality of lists includes:

locating the plurality of lists within a plurality of documents on one or more networks.

5. The method of claim 1, wherein the plurality of lists are identified by tags, by items stored in a table, or by items separated by tabs, commas, or semicolons.

6. The method of claim 1, further comprising:

creating a hit list index from the plurality of lists, the hit list index providing a mapping of items to one or more of the plurality of lists in which the items appear.

7. The method of claim 1, wherein the plurality of lists are stored in one or more databases.

8. The method of claim 1, wherein determining the probabilities includes:

determining a probability of one of the plurality of lists being generated using the on-topic model based on a probability of each of the items in the one of the plurality of lists being generated from the on-topic model.

9. The method of claim 1, further comprising:

assigning weights to the items in one of the plurality of lists based on the determined probability associated with the one of the plurality of lists; and

adding the weights for each of the items to generate total weights for the items.

10. The method of claim 9, wherein the forming a list includes:

forming the list based on the items from the plurality of lists and the total weights for the items.

11. The method of claim 9, further comprising:

updating the on-topic model based on the items and the total weights for the items; and

determining new probabilities for the plurality of lists based on the updated on-topic model.

12. The method of claim 11, wherein the forming a list includes:

forming the list based on the plurality of lists and the new probabilities.

13. The method of claim 11, further comprising:

repeating the updating of the on-topic model and the determining of the new probabilities for the plurality of lists for a predetermined number of iterations.

14. The method of claim 1, wherein the probabilities of the plurality of lists being generated using the on-topic model are related to a number of items in the plurality of lists.

15. The method of claim 1, wherein the list includes a plurality of selectable items, selection of one of the selectable items causing a search of documents relating to the one selectable item to be performed.

16. The method of claim 1, further comprising:

determining an off-topic model.

17. The method of claim 16, wherein determining the probabilities includes:

determining probabilities for the plurality of lists based on the on-topic model or the off-topic model.

18. The method of claim 16, further comprising:

receiving one or more off-topic example items; and

wherein the determining an off-topic model includes;

determining a probabilistic model based on the one or more off-topic example items.

19. A system, comprising:

means for identifying a plurality of lists associated with one or more documents;

means for determining an on-topic model and an off-topic model;

means for classifying the plurality of lists based on the on-topic model and the off-topic model to determine probabilities of the plurality of lists being generated using the on-topic model or the off-topic model;

means for forming a list from items in the plurality of lists based on the determined probabilities; and

means for providing the list to a client.

20. A system, comprising:

a list identifier configured to identify a plurality of lists by crawling one or more networks, each of the plurality of lists including a plurality of items;

a list classifier configured to;

determine an on-topic model, and

determine confidence scores reflecting measures of confidence of the plurality of lists being generated using the on-topic model; and

a list processor configured to;

form a list from the items in the plurality of lists based on the determined confidence scores associated with the plurality of lists, and

present the list to a client.

21. A system, comprising:

means for receiving one or more example items;

means for identifying a plurality of lists by crawling one or more networks;

means for determining confidence scores for the plurality of lists based on the one or more example items, each of the plurality of lists including a plurality of items;

means for forming a list from the items in the plurality of lists based on the confidence scores for the plurality of lists; and

means for providing the list to a client.

22. A system, comprising:

a list identifier configured to identify a plurality of lists, each of the plurality of lists including a plurality of items;

a list classifier configured to assign confidence weights to the plurality of lists;

and

a list processor configured to;

form a list from the items in the plurality of lists based on the confidence weights assigned to the plurality of lists, and

present the list to a client.

23. A system, comprising:

a list identifier configured to identify a plurality of lists, each of the plurality of lists including a plurality of items;

a list classifier configured to;

receive one or more example items,

determine probabilities that the plurality of lists are associated with the one or more example items,

assign weights to the items in the plurality of lists based on the probabilities of the plurality of lists, and

add the weights for each of the items to generate total weights for the items; and

a list processor configured to;

form a list based on the items and the total weights for the items, and

present the list to a client.

24. A method, comprising:

receiving one or more example items;

determining an on-topic model based on the one or more example items;

classifying a plurality of lists based on the on-topic model;

assigning weights to items in the plurality of lists based on the classifying of the plurality of lists;

adding the weights for each of the items to generate a total weight for each of the items;

forming a list based on the items and the total weights for the items; and

presenting the list to a client.

25. The method of claim 24, further comprising:

updating the on-topic model based on the items and the total weights for the items;

reclassifying the plurality of lists based on the updated on-topic model;

reassigning weights to the items based on the reclassifying of the plurality of lists; and

adding the reassigned weights for each of the items to generate a new total weight for each of the items.

26. The method of claim 25, wherein the forming the list includes:

forming the list based on the items and the new total weights for the items.

27. A system, comprising:

a master configured to;

receive, from a client, one or more example items, and

determine an on-topic model based on the one or more example items; and

a plurality of slaves, each of the slaves being configured to;

classify a subset of a plurality of lists based on the on-topic model,

assign weights to items in the plurality of lists based on the classified lists,

add the weights for each of the items to generate a total weight for each of the items,

send the items and the total weights for the items to the master,

the master using the items and the total weights for the items to form a list, and present the list to the client.