Non-text content item search

• US 8,856,125 B1
• Filed: 09/07/2012
• Issued: 10/07/2014
• Est. Priority Date: 02/26/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Malpani, Radhika, Preetham, Arcot J., Mate, Omkar
• Primary Examiner(s): Vy, Hung T
• Application Number:
  US13/606,336
• Time in Patent Office: 760 Days
• Field of Search: 707/737, 707/748, 707/738, 707/749, 707/776, 382/118, 382/173
• US Class Current: 707/737
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for selecting labels for a non-text content item. In one aspect, a method receives a set of initial labels for a non-text content item, wherein the set of initial labels specifies text that has been identified as descriptive of the non-text content item and a web page to which the text corresponds. Initial labels corresponding to sets of matching web pages are grouped into separate initial label groups that correspond to each set of matching web pages. Sets of matching labels are grouped into other separate initial label groups that correspond to the sets of matching labels. One or more words that are included in at least a threshold number of the separate label groups are selected as final labels for the non-text content item.

Claims

1. A method performed by data processing apparatus, the method comprising:

identifying a non-text content item that is associated with each of a plurality of web pages;

receiving label data that includes a set of initial labels for the non-text content item, wherein each initial label includes one or more words;

grouping, for each of two or more sets of matching web pages among the plurality of web pages, initial labels that are associated with the set of matching web pages into a label group, the initial labels for different set of matching web pages being grouped to different label groups;

grouping different sets of matching labels from the set of initial labels into different label groups; and

selecting, as a final label for the non-text content item, an n-gram of one or more words that is included in at least a threshold number of different label groups.

2. The method of claim 1, wherein selecting a final label for the non-text content item comprises:

identifying one or more n-grams for each of the different label groups, each n-gram including one or more words from initial labels that are included in the different label group;

determining, for each identified n-gram, a number of the different label groups for which the n-gram was identified; and

selecting, as a final label for the non-text content item, an identified n-gram for which the number of different label groups meets the threshold number of different label groups.

3. The method of claim 1, wherein grouping initial labels that are associated with the set of matching web pages into a label group comprises:

identifying, among the set of initial labels, a first set of at least two initial labels that were obtained from a same domain;

grouping the first set of at least two initial labels into a label group;

identifying, among the set of initial labels, a second set of at least two initial labels that were obtained from a same domain; and

grouping the second set of at least two initial labels into a label group.

4. The method of claim 1, wherein grouping different sets of matching labels from the set of initial labels into different label groups comprises:

identifying, from among the set of initial labels, a first set of at least two initial labels that have at least a threshold measure of similarity;

grouping the first set of at least two initial labels into a first label group;

identifying, from among the set of initial labels, a second set of at least two initial labels that have at least a threshold measure of similarity; and

grouping the second set of at least two initial labels into a second label group.

5. The method of claim 1, further comprising:

identifying one or more sets of matching web pages from the plurality of web pages, each set of matching web pages being a two or more web pages that each have a characteristic that matches a characteristic of another web page in the set of matching web pages; and

identifying one or more sets of matching labels from the set of initial labels, each set of matching labels including two or more initial labels having at least a threshold measure of similarity.

6. The method of claim 5, wherein identifying one or more sets of matching web pages comprises identifying at least one set of matching web pages in which web pages are associated with a same domain.

7. The method of claim 5, wherein identifying one or more sets of matching labels comprises identifying at least one set of matching labels in which at least two labels each include a same word.

8. The method of claim 5, wherein identifying one or more sets of matching labels comprises identifying at least one set of matching labels in which each label references a same topic.

9. A non-transitory computer storage medium encoded with a computer program, the program comprising instructions that when executed by data processing apparatus cause the data processing apparatus to perform operations comprising:

identifying a non-text content item that is associated with each of a plurality of web pages;

receiving label data that includes a set of initial labels for the non-text content item, wherein each initial label includes one or more words;

grouping, for each of two or more sets of matching web pages among the plurality of web pages, initial labels that are associated with the set of matching web pages into a label group, the initial labels for different set of matching web pages being grouped to different label groups;

grouping different sets of matching labels from the set of initial labels into different label groups; and

selecting, as a final label for the non-text content item, an n-gram of one or more words that is included in at least a threshold number of different label groups.

10. The non-transitory computer storage medium of claim 9, wherein selecting a final label for the non-text content item comprises:

identifying one or more n-grams for each of the different label groups, each n-gram including one or more words from initial labels that are included in the different label group;

determining, for each identified n-gram, a number of the different label groups for which the n-gram was identified; and

selecting, as a final label for the non-text content item, an identified n-gram for which the number of different label groups meets the threshold number of different label groups.

11. The non-transitory computer storage medium of claim 9, wherein grouping initial labels that are associated with the set of matching web pages into a label group comprises:

identifying, among the set of initial labels, a first set of at least two initial labels that were obtained from a same domain;

grouping the first set of at least two initial labels into a label group;

identifying, among the set of initial labels, a second set of at least two initial labels that were obtained from a same domain; and

grouping the second set of at least two initial labels into a label group.

12. The non-transitory computer storage medium of claim 9, wherein grouping different sets of matching labels from the set of initial labels into different label groups comprises:

identifying, from among the set of initial labels, a first set of at least two initial labels that have at least a threshold measure of similarity;

grouping the first set of at least two initial labels into a first label group;

identifying, from among the set of initial labels, a second set of at least two initial labels that have at least a threshold measure of similarity; and

grouping the second set of at least two initial labels into a second label group.

13. The non-transitory computer storage medium of claim 9, wherein the instructions further cause the data processing apparatus to perform operations comprising:

identifying one or more sets of matching web pages from the plurality of web pages, each set of matching web pages being a two or more web pages that each have a characteristic that matches a characteristic of another web page in the set of matching web pages; and

identifying one or more sets of matching labels from the set of initial labels, each set of matching labels including two or more initial labels having at least a threshold measure of similarity.

14. The non-transitory computer storage medium of claim 13, wherein identifying one or more sets of matching web pages comprises identifying at least one set of matching web pages in which web pages are associated with a same domain.

15. The non-transitory computer storage medium of claim 13, wherein identifying one or more sets of matching labels comprises identifying at least one set of matching labels in which at least two labels each include a same word.

16. The non-transitory computer storage medium of claim 13, wherein identifying one or more sets of matching labels comprises identifying at least one set of matching labels in which each label references a same topic.

17. A system comprising:

a data store storing label data that specifies a set of initial labels for a non-text content item, wherein the non-text content item is associated with each of a plurality of web pages, and wherein each initial label includes one or more words; and

one or more computers coupled to the data store, the one or more computers storing instructions that cause the one or more computers to interact with the data store and perform operations comprising;

identifying a non-text content item that is associated with each of a plurality of web pages;

receiving label data that includes a set of initial labels for the non-text content item, wherein each initial label includes one or more words;

grouping, for each of two or more sets of matching web pages among the plurality of web pages, initial labels that are associated with the set of matching web pages into a label group, the initial labels for different set of matching web pages being grouped to different label groups;

grouping different sets of matching labels from the set of initial labels into different label groups; and

selecting, as a final label for the non-text content item, an n-gram of one or more words that is included in at least a threshold number of different label groups.

18. The system of claim 17, wherein the instructions cause the one or more computers to select a final label for the non-text content item by performing operations comprising:

identifying one or more n-grams for each of the different label groups, each n-gram including one or more words from initial labels that are included in the different label group;

determining, for each identified n-gram, a number of the different label groups for which the n-gram was identified; and

selecting, as a final label for the non-text content item, an identified n-gram for which the number of different label groups meets the threshold number of different label groups.

19. The system of claim 17, wherein the instructions cause the one or more computers to group initial labels that are associated with the set of matching web pages into a label group by performing operations comprising:

identifying, among the set of initial labels, a first set of at least two initial labels that were obtained from a same domain;

grouping the first set of at least two initial labels into a label group;

identifying, among the set of initial labels, a second set of at least two initial labels that were obtained from a same domain; and

grouping the second set of at least two initial labels into a label group.

20. The system of claim 17, wherein the instructions cause the one or more computers to group different sets of matching labels from the set of initial labels into different label groups by performing operations comprising:

identifying, from among the set of initial labels, a first set of at least two initial labels that have at least a threshold measure of similarity;

grouping the first set of at least two initial labels into a first label group;

identifying, from among the set of initial labels, a second set of at least two initial labels that have at least a threshold measure of similarity; and

grouping the second set of at least two initial labels into a second label group.

21. The system of claim 17, wherein the instructions further cause the one or more computers to perform operations comprising:

identifying one or more sets of matching web pages from the plurality of web pages, each set of matching web pages being a two or more web pages that each have a characteristic that matches a characteristic of another web page in the set of matching web pages; and

identifying one or more sets of matching labels from the set of initial labels, each set of matching labels including two or more initial labels having at least a threshold measure of similarity.

22. The system of claim 21, wherein the instructions cause the one or more computers to identify one or more sets of matching web pages by performing operations comprising identifying at least one set of matching web pages in which web pages are associated with a same domain.

23. The system of claim 21, wherein the instructions cause the one or more computers to identify one or more sets of matching labels by causing the one or more computers to perform operations comprising identifying at least one set of matching labels in which at least two labels each include a same word.

24. The system of claim 21, wherein the instructions cause the one or more computers to identify one or more sets of matching labels by causing the one or more computers to perform operations comprising identifying at least one set of matching labels in which each label references a same topic.