Enhancing search results using conceptual document relationships

• US 8,090,736 B1
• Filed: 12/30/2004
• Issued: 01/03/2012
• Est. Priority Date: 12/30/2004
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Knapp, Sean Michael, Waddell, Matthew Eric
• Primary Examiner(s): Hwa, Shyue Jiunn
• Application Number:
  US11/024,966
• Time in Patent Office: 2,560 Days
• Field of Search: 707/760, 707/999.004
• US Class Current: 707/760
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06Q 20/208 Input by product or record sensing, e.g. weighing or scanner processing

Abstract

A system identifies user-defined organizational relationships between documents in a collection of documents. The system ranks the documents based on the identified relationships. In response to a user search query, a listing of matching documents is identified. The listing of matching document is presented to the user based on document ranks.

Claims

1. A method performed by a computer system, the method comprising:

selecting, using one or more processors associated with the computer system, a first document;

representing, using one or more processors associated with the computer system, a first relationship between the first document and a second document as a first link between the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user-defined label,

the first document and the second document being associated with a same sub-directory, or

the first document and the second document being associated with a same bookmark folder;

representing, using one or more processors associated with the computer system, a second relationship between the first document and a third document as a second link between the first document and a third document, where the second relationship includes at least one of;

a hierarchical user-defined label associated with the first document and a hierarchical user-defined label associated with the third document being sibling labels,

a sub-directory associated with the first document and a sub-directory associated with the third document being sibling sub-directories, or

a bookmark folder associated with the first document and a bookmark folder associated with the third document being sibling bookmark folders; and

assigning, using one or more processors associated with the computer system, a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link.

2. The method of claim 1, where the hierarchical user-defined labels are assigned to the first document by a number of unique users.

3. The method of claim 1, comprising:

representing a third relationship between the second document and a fourth document as a third link between the second document and the fourth document, where the third relationship includes the second document and the fourth document sharing a same hierarchical user-defined label that is not shared by the first document; and

modifying the rank score of the first document based on the third link and based on a rank score associated with the fourth document.

4. The method of claim 3, where the fourth document comprises a number of intermediary documents, where each of the intermediary documents shares at least one hierarchical user-defined label with at least one other intermediary document.

5. The method of claim 3, where the third link and a rank score associated with the fourth document have less effect on the rank score of the first document than the first link.

6. The method of claim 1, where the rank score assigned to the first document is based on a rank score of the second document and a rank score of the third document.

7. The method of claim 1, further comprising:

determining, for the first document, a number of other documents that share the same user-hierarchical label with the first document, are associated with the same sub-directory as the first document, or are associated with the same bookmark folder as the first document; and

where assigning the rank score to the first document is further based on the number of other documents.

8. The method of claim 1, further comprising:

receiving a search query from a user;

identifying the first document based on the received search query;

calculating an information retrieval score for the first document;

calculating a total score for the first document based on the information retrieval score and the assigned rank score; and

providing, for display, the first document together with other identified documents to the user, where the position of the first document among the other identified documents is based on the total score.

9. A method performed by a computer system, the method comprising:

receiving, using a communication interface associated with the computer system, labels for a first document from a plurality of different users;

detecting, using one or more processors associated with the computer system, a first link between the first document and a second document, based on at least one of the received labels matching a label associated with the second document;

detecting, using one or more processors associated with the computer system, a second link between the first document and a third document, based on at least one of the received labels and a label associated with the third document being sibling labels;

assigning, using one or more processors associated with the computer system, a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link, and where the rank score is further based on a quantity of the different users;

detecting, using one or more processors associated with the computer system, a third link between the second document and a fourth document, based on the second document and the fourth document sharing a same label that is not shared by the first document; and

modifying, using one or more processors associated with the computer system, the rank score of the first document based on the detected third link.

10. The method of claim 9, were the rank score is further based on a quantity of discrete labels received for the first document from the plurality of different users.

11. The method of claim 9, further comprising:

receiving a search query;

selecting a group of matching documents from a collection of documents based on the search query, where the group of matching documents includes the first document; and

presenting one or more documents of the group of matching documents, where a position of the first document, among the presented one or more documents, is based on the rank score assigned to the first document.

12. A search system, comprising:

one or more computer devices to;

represent a first relationship between the first document and a second document as a first link between the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user- defined label,

the first document and the second document being associated with a same sub- directory, or

the first document and the second document being associated with a same bookmark folder;

represent a second relationship between the first document and the third document as a second link between the first document and a third document, where the second relationship includes at least one of;

a hierarchical user-defined label, associated with the first document, and a hierarchical user-defined label, associated with the third document, being sibling labels,

a sub-directory, associated with the first document, and a sub-directory, associated with the third document, being sibling sub-directories, or

a bookmark folder, associated with the first document, and a bookmark folder, associated with the third document, being sibling bookmark folders;

assign a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link;

representing, using one or more processors associated with the one or more computer devices, a second relationship between the first document and a third document as a second link between the first document and the third document, where the second relationship includes at least one of;

a hierarchical user-defined label, associated with the first document, and a hierarchical user-defined label, associated with the third document, being sibling labels,

a sub-directory, associated with the first document, and a sub-directory, associated with the third document, being sibling sub-directories, or

a bookmark folder, associated with the first document, and a bookmark folder, associated with the third document, being sibling bookmark folders;

assigning, using one or more processors associated with the one or more computer devices, a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link;

receiving, using one or more processors associated with the one or more computer devices, a search query;

generating, using one or more processors associated with the one or more computer devices, an information retrieval score for the first document based on the search query;

generating, using one or more processors associated with the one or more computer devices, a total rank score for the first document based on the assigned rank score and based on the generated information retrieval score; and

receive a search query;

select a group of matching documents from a collection of documents based on the search query, where the group of matching documents includes the first document; and

present one or more documents of the group of matching documents, where a position of the first document, among the presented one or more documents, is based on the rank score assigned to the first document.

13. The search system of claim 12, where the one or more server devices are further to:

present a user interface for receiving the search query, where the user interface includes hyperlinks relating hierarchical user-defined labels, bookmarks, or sub-directories associated with a particular document.

14. The search system of claim 13, where the user interface further includes a hyperlink to allow a user to edit the hierarchical user-defined labels, bookmarks, or sub-directories.

15. One or more memory devices that stores instructions executable by one or more processors, the one or more memory devices comprising:

one or more instructions to receive labels for a first document from a plurality of different users;

one or more instructions to detect a first link between the first document and a second document, based on at least one of the received labels matching a label associated with the second document;

one or more instructions to detect a second link between the first document and a third document, based on at least one of the received labels and a label, associated with the third document, being sibling labels;

one or more instructions to assign a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link, and where the rank score is further based on a quantity of the different users;

one or more instructions to detect a third link between the second document and a fourth document, based on the second document and the fourth document sharing a same label that is not shared by the first document; and

one or more instructions to modify the rank score of the first document based on the detected third link.

16. The one or more memory devices of claim 15, where the rank score for the first document is further based on rank scores associated with other documents associated with labels that match the at least one of the received labels.

17. The one or more memory devices of claim 16, where the rank score for the first document is based on a sum of values, where each value is a ratio of a rank score associated with one of the other documents and a number of documents sharing a label with the one of the other documents.

18. A method performed by one or more computer devices, the method comprising:

retrieving, using one or more processors associated with the one or more computer devices, hierarchical user-defined labels assigned to a first document;

representing, using one or more processors associated with the one or more computer devices, a first relationship between the first document and a second document as a first link between the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user-defined label,

the first document and the second document being associated with a same sub-directory, or

the first document and the second document being associated with a same bookmark folder;

presenting, using one or more processors associated with the one or more computer devices, one or more documents selected from a collection of documents, where the one or more presented documents include the first document and where a position of the first document among the one or more presented documents is based on the generated total rank score.

19. The method of claim 18, where the presenting the one or more documents comprises presenting references for the one or more documents, where the references include a title and a text excerpt from the particular document.

20. The method of claim 18, where the assigning a rank score comprises increasing the rank score of the first document if the first document has hierarchical user-defined labels assigned to the first document by a plurality of different users.

21. A system comprising:

one or more computer devices to;

select a first document;

represent a first relationship between the first document and a second document as a first link between the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user-defined label,

the first document and the second document being associated with a same sub-directory, or

the first document and the second document being associated with a same bookmark folder;

represent a second relationship between the first document and a third document as a second link between the first document and the third document, where the second relationship includes at least one of;

a hierarchical user-defined label associated with the first document and a hierarchical user-defined label associated with the third document being sibling labels,

a sub-directory associated with the first document and a sub-directory associated with the third document being sibling sub-directories, or

a bookmark folder associated with the first document and a bookmark folder associated with the third document being sibling bookmark folders; and

assign a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link.

22. The system of claim 21, where the hierarchical user-defined label is assigned to the first document by a number of unique users.

23. The system of claim 21, where the rank score assigned to the first document is based on a rank score of the second document and a rank score of the third document.

24. The system of claim 21, where the one or more computer devices are further to:

represent a third relationship between the second document and a fourth document as a third link between the second document and the fourth document, where the third relationship includes the second document and the fourth document sharing a same hierarchical user-defined label that is not shared by the first document; and

modify the rank score of the first document based on the third link and based on a rank score associated with the fourth document.

25. The system of claim 24, where the fourth document is one of a number of intermediary documents, where each of the intermediary documents shares at least one hierarchical user-defined label with at least one other intermediary document.

26. The system of claim 21, where the one or more computer devices are further to:

determine, for the first document, a number of other documents that share the same user-hierarchical label with the first document, are associated with the same sub-directory as the first document, or are associated with the same bookmark folder as the first document; and

where the rank score assigned to the first document is further based on the number of other documents.

27. The system of claim 21, where the one or more computer devices are further to:

receive a search query from a user;

identify the first document based on the received search query;

calculate an information retrieval score for the first document;

calculate a total score for the first document based on the information retrieval score and the assigned rank score; and

provide, for display, the first document together with other identified documents to the user, where a position of the first document among the other identified documents is based on the total score.

28. A system comprising:

one or more computer devices to;

receive labels for a first document from a plurality of different users;

detect a first link between the first document and a second document, based on at least one of the received labels matching a label associated with the second document;

detect a second link between the first document and a third document, based on at least one of the received labels and a label associated with the third document being sibling labels;

assign a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link, and where the rank score is further based on a quantity of the plurality of different users;

detect a third link between the second document and a fourth document, based on the second document and the fourth document sharing a same label that is not shared by the first document; and

modify the rank score of the first document based on the detected third link.

29. The system of claim 28, were the rank score is further based on a quantity of discrete labels received for the first document from the plurality of different users.

30. The system of claim 28, where the one or more computer devices are further to:

receive a search query;

select a group of matching documents from a collection of documents based on the search query, where the group of matching documents includes the first document; and

present one or more documents of the group of matching documents, where a position of the first document, among the presented one or more documents, is based on the rank score assigned to the first document.

31. A method performed by one or more computer devices, the method comprising:

representing, by at least one of the one or more computer devices, a first relationship between the first document and a second document as a first link between

the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user-defined label,

the first document and the second document being associated with a same sub-directory, or

the first document and the second document being associated with a same bookmark folder;

representing, by at least one of the one or more computer devices, a second relationship between the first document and the third document as a second link between the first document and a third document, where the second relationship includes at least one of;

a hierarchical user-defined label, associated with the first document, and a hierarchical user-defined label, associated with the third document, being sibling labels,

a sub-directory, associated with the first document, and a sub-directory, associated with the third document, being sibling sub-directories, or

a bookmark folder, associated with the first document, and a bookmark folder, associated with the third document, being sibling bookmark folders;

assigning, by at least one of the one or more computer devices, a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link;

receiving, by at least one of the one or more computer devices, a search query;

selecting, by at least one of the one or more computer devices, a group of matching documents from a collection of documents based on the search query, where the group of matching documents includes the first document; and

presenting, by at least one of the one or more computer devices, one or more documents of the group of matching documents, where a position of the first document, among the presented one or more documents, is based on the rank score assigned to the first document.

32. The method of claim 31, further comprising:

presenting a user interface for receiving the search query, where the user interface includes hyperlinks relating hierarchical user-defined labels, bookmarks, or sub-directories associated with a particular document.

33. The method of claim 32, where the user interface further includes a hyperlink to allow a user to edit the hierarchical user-defined labels, bookmarks, or sub-directories.

34. A system comprising:

one or more computer devices to;

retrieve hierarchical user-defined labels assigned to a first document;

represent a first relationship between the first document and a second document as a first link between the first document and the second document, where the first relationship includes at least one of;

the first document and the second document sharing a same hierarchical user-defined label,

the first document and the second document being associated with a same sub-directory, or

the first document and the second document being associated with a same bookmark folder;

represent a second relationship between the first document and a third document as a second link between the first document and the third document, where the second relationship includes at least one of;

a hierarchical user-defined label, associated with the first document, and a hierarchical user-defined label, associated with the third document, being sibling labels,

a sub-directory, associated with the first document, and a sub-directory, associated with the third document, being sibling sub-directories, or

a bookmark folder, associated with the first document, and a bookmark folder, associated with the third document, being sibling bookmark folders;

assign a rank score to the first document based on the first link and based on the second link, where the first link contributes more to the rank score than the second link;

receive a search query;

generate an information retrieval score for the first document based on the search query;

generate a total rank score for the first document based on the assigned rank score and based on the generated information retrieval score; and

present one or more documents selected from a collection of documents, where the one or more presented documents include the first document and where a position of the first document among the one or more presented documents is based on the generated total rank score.

35. The system of claim 34, where, when the one or more computer devices are to present the one or more documents, the one or more computer devices are further to:

present references for the one or more documents, where the references include a title and a text excerpt from the one or more documents.

36. The system of claim 34, where, when the one or more computer devices are to assign a rank score, the one or more computer devices are to:

increase the rank score of the first document if the first document has hierarchical user-defined labels assigned to the first document by a plurality of different users.