Building and using subwebs for focused search

• US 20050165753A1
• Filed: 02/13/2004
• Published: 07/28/2005
• Est. Priority Date: 01/23/2004
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Brill, Eric D., Chandrasekar, Raman, Chen, Harr, Corston, Simon H.
• Granted Patent: US 7,392,278 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9536 Search customisation based on social or collaborative filtering
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A system that facilitates performance of a focused search over a collection of sites comprises a subweb that corresponds to a topic and/or user characteristic(s) that are of interest to the user. The subweb includes a plurality of domains and/or paths (e.g. sites) that are related to the topic and/or the user characteristic(s). Each of the sites within the subweb is assigned a weight that indicates relevance of the site to the desirable topic and/or user characteristic(s). A search engine employs the subweb to facilitate focusing a search over a collection of sites. The search engine receives a query, and utilizes the subweb to focus a search over the selection of sites corresponding to the topic and/or user characteristic(s) represented by the subweb. The results from the search are returned to the user based at least in part upon the relevance weights assigned to the sites within the subweb.

Claims

1. A system that facilitates searching, comprising:

a display component that presents a plurality of subwebs to search on over a collection of items, the subwebs respectively are associated with various subsets of the items weighted by relevance; and

an input component that receives a search query over at least one of the subwebs.

2. The system of claim 1 resides on a server.

3. The system of claim 1 resides on a client.

4. The system of claim 1, the subwebs are personalized to users of the system.

5. The system of claim 4, the system further comprising a machine-learning component that builds profiles in connection with the users.

6. The system of claim 1, further comprising a context component that presents the plurality of subwebs as a function of user context.

7. The system of claim 5, the profiles are transportable.

8. A crawling system, comprising:

a crawling component that crawls sites as a function of pre-determined subwebs associated with items, the subwebs respectively are associated with various subsets of the items weighted by relevance.

9. The system of claim 8, further comprising a utility component that performs a probabilistic-based analysis in connection with the crawling.

10. The system of claim 9, the utility component performs a cost-benefit analysis in connection with the crawling.

11. The system of claim 8, the crawling component being a web crawler that crawls at least one of the Internet and an intranet.

12. A system that facilitates performance of a focused search over a collection of sites, comprising:

a subweb, the subweb corresponding to at least one of a particular topic and user characteristic(s), the subweb comprising a plurality of sites related to the at least one of the topic and the user characteristic(s), each of the plurality of sites assigned a relevance weight that indicates relevance of the sites to the at least one of the topic and the user characteristic(s); and

a search component that receives a query, the search component employing the subweb to focus a search over the collection of sites based upon the query, the search component returns results based at least in part upon the subweb and the query.

13. The system of claim 12, further comprising a subweb selector that selects a subweb to employ in connection with the search.

14. The system of claim 13, the subweb selector selects a subweb based at least in part upon point-of-entry on the collection of sites.

15. The system of claim 13, the subweb selector selects a subweb based at least in part upon one of the query, user state, user context, and user history.

16. The system of claim 15, the subweb selector selects a subweb based at least in part upon a sensed user location.

17. The system of claim 15, the subweb selector selects a subweb based at least in part upon one or more of time of a day and a day of a week.

18. The system of claim 13, the subweb selector associated with a user interface, wherein a user selects a subweb to employ in connection with the search via the user interface.

19. The system of claim 12, further comprising a tuning component that facilitates one of adding and removing a site from the subweb.

20. The system of claim 12, further comprising a tuning component that alters an assigned relevance weight of one or more of the plurality of sites within the subweb.

21. The system of claim 20, the tuning component associated with a user interface, wherein a user alters an assigned relevance weight to the one or more of the plurality of sites within the subweb via the user interface.

22. The system of claim 20, further comprising crawling component that crawls sites as a function of the subweb, the crawling component and the tuning component operate conjunctively to automatically maintain the subweb.

23. The system of claim 12, the subweb accessible via a web site.

24. The system of claim 12, the subweb associated with a user profile, the user profile being portable.

25. The system of claim 12, further comprising a component that accepts feedback relating to the results, the feedback employed to automatically maintain the subweb.

26. The system of claim 12, an order of the results based upon a function of a general search engine ranking system and the assigned weights of the plurality of sites within the subweb.

27. The system of claim 12, wherein at least one of the results returned that does not match a site within the subweb is deleted from the results.

28. The system of claim 12, the user characteristic(s) related to one or more of age, location, sex, ethnicity, political affiliation, and religion.

29. The system of claim 12, wherein a plurality of subwebs are utilized in connection with the search and results obtained from the search are classified according to a subweb related to the results.

30. The system of claim 12, the collection of sites being at least one of the Internet and an intranet.

31. A method for performing a focused search, comprising:

providing a subweb that is defined by a plurality of sites with relevance to at least one of a topic and user characteristic(s) represented by the subweb, the sites assigned a relevance weight that indicates each site'"'"'s relevance to the at least one of the topic and the user characteristic(s);

selecting the subweb;

relaying a query related to the at least one of the topic and the user characteristic(s) represented by the subweb to a search engine;

searching a collection of sites for information based upon the query;

obtaining search results based at least in part upon the query; and

ranking the search results based at least in part by the relevance weight assigned to the sites that define the subweb.

32. The method of claim 31, further comprising deleting at least one search result that is not associated with the sites that define the subweb.

33. The method of claim 31, further comprising selecting the subweb based at least in part upon at least one of the query, user history, user context, and user state.

34. The method of claim 31, the collection of sites being at least one of the Internet and an intranet.

35. The method of claim 34, further comprising selecting the subweb based upon an entry point on the at least one of the Internet and the intranet.

36. The method of claim 31, further comprising employing a crawler to automatically maintain the subweb.

37. The method of claim 31, further comprising altering at least one relevance weight based at least in part upon user feedback in connection with the search results.

38. The method of claim 31, further comprising altering at least one relevance weight based at least in part upon one or more of the query, user history, user state, and user context.

39. A subweb building system, comprising:

a site locator that locates sites within a collection of sites relating to one or more of a particular topic and a user characteristic(s);

a relevance weight assignor that assigns a relevance weight to a located site according to the site'"'"'s relevance to the one or more of the topic and the user characteristic(s).

40. The system of claim 39, the site locator comprising:

a list of queries that are associated with the one or more of the topic and the user characteristic(s); and

a search engine that searches the collection of sites based upon the list of queries and obtains a list of results.

41. The system of claim 40, the relevance weight assignor comprising:

a path distribution calculator that computes a distribution path of the results;

a neighborhood path distribution calculator that computes a distribution of sites neighboring sites returned in the results; and

a net path distribution calculator that computes a net path distribution.

42. The system of claim 41, the weight assignor further comprising a filtering component that removes non-related sites from the results.

43. The system of claim 42, the non-related sites removed from the results at least in part via providing the search engine with a plurality of random queries and comparing results from the random queries with results from the queries that are associated with one or more of the topic and the user characteristic(s).

44. The system of claim 43 employed by a client.

45. The system of claim 39, at least one of the Internet and an intranet being the collection of sites.

46. The system of claim 39, further comprising a stop list, the stop list comprising a list of sites that are known a priori to be irrelevant to the one or more of the topic and the user characteristic(s), the list of sites within the stop list prevented from being within the collection of sites related to the one or more of the topic and the user characteristic(s).

47. A method for generating a subweb, comprising:

collecting a plurality of sites related to at least one of a topic and a user characteristic, the sites defining the subweb; and

assigning biased relevance weight to a subset of the plurality of sites.

48. The method of claim 47, further comprising:

providing a user interface; and

entering sites related to the topic via the user interface.

49. The method of claim 47, further comprising:

providing a plurality of queries related to the at least one of the topic and the user characteristic(s);

providing a search engine;

relaying the plurality of queries to the search engine and performing a search over a collection of sites; and

caching results obtained from the search.

50. The method of claim 49, further comprising:

computing a path distribution of the results of the search; and

computing a path distribution of sites that neighbor sites within the results, wherein neighboring sites includes inlinked and outlinked sites;

computing a path distribution of the results of the search and the neighboring sites; and

normalizing the path distribution of the results of the search and the neighboring sites.

51. The method of claim 50, further comprising:

providing a plurality of random queries;

relaying the queries to the general search engine and performing a search over a collection of sites; and

normalizing the path distribution of the results of the search and the neighboring sites based at least in part upon the results obtained from the search based upon the random queries.

57. The system of claim 51, the collection of sites being at least one of the Internet and an intranet.

52. A system for searching a collection of sites, comprising:

means for generating a topic-specific subweb, the topic-specific subweb comprising a plurality of sites related to the topic, the sites assigned a weight according to relevance of the sites to the topic; and

means for employing the subweb in connection with a search engine to search the collection of sites.

53. The system of claim 52, further comprising means for assigning the weights to the sites within the subweb.

54. The system of claim 52, further comprising means for tuning the subweb.

55. The system of claim 52, further comprising means for selecting the subweb from a plurality of subwebs.

56. The system of claim 52, further comprising means for displaying results of a search corresponding to a subweb that boosts the results.

58. A computer readable medium having computer executable instructions stored thereon to:

collect a plurality sites that are relevant to at least one of a topic and user characteristic(s);

assign weights to each of the plurality of sites corresponding to calculated level of relevance of the sites to the at least one of the topic and the user characteristic(s); and

perform a search over a subset of the plurality of sites based upon a received query.