Personalized query completion suggestion

• US 8,027,964 B2
• Filed: 07/13/2007
• Issued: 09/27/2011
• Est. Priority Date: 07/13/2007
• Status: Active Grant

• Current Assignee: HERE Global B.V. (Nokia Corporation)
• Original Assignee: Medio Systems, Inc. (Nokia Corporation)
• Inventors: Boulis, Konstantinos
• Primary Examiner(s): Lee; Wilson
• Assistant Examiner(s): Le; Jessica N
• Application Number:
  US11/777,977
• Publication Number:
  US 20090019002A1
• Time in Patent Office: 1,537 Days
• Field of Search: None
• US Class Current: 707/705
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 9/451 Execution arrangements for user interfaces

Abstract

Search query processing includes receiving a search query input string from a user of a mobile device and comparing the search query input to a personalized dictionary of the mobile device, determining a suggested completion for each match in the comparison, and providing the suggested completion to the user for selection. The user can select the suggested modification, if desired, and the completed query can be submitted to a search facility. Personalized dictionaries can be generated from analysis of previously submitted search queries. The analyzed search queries may have been submitted by the user, as well as by other users with similar interests. The analysis of search queries may categorize users into clusters or groups of persons having similar interests.

Claims

1. A computer system for generating a personalized search dictionary for query completion suggestion, the system comprising:

a user profile creating component comprising a processor of the computer system configured to perform analysis of raw search query information for each user of a user population and creates a user profile for each user, wherein the user profile identifies the user and corresponding raw search queries submitted by the identified user over a period of time;

a query processing and user token profile component comprising a processor of the computer system configured to process the user profiles and perform query information processing, including spelling correction, query tokenizing, and creation of user-token profiles, in which search query terms of the raw search queries in each of the user profiles from the user profile creating component are parsed into tokens comprising letters and identifiable words and a user-token profile corresponding to each user profile is created comprising the raw search queries replaced with the tokenized search queries;

a search dictionary component comprising a processor of the computer system for defining one or more groups based on the user-token profiles wherein each user of the user population is assigned to one or more of the defined groups, wherein each defined group comprises a collection of users that are indicated by frequency of the tokenized search queries as having similar preferences and generates a search dictionary for each defined group in accordance with the tokenized search queries of the users in the corresponding defined group;

wherein the search dictionary component generates a personalized search dictionary for the user in accordance with data submitted by a predetermined user population over a period of time wherein the personalized search dictionary provides the user with query completion suggestions that are configured to be submitted to a search facility in response to being accepted by the user.

2. The system as in claim 1, wherein the search dictionary component assigns a target user into one or more of the identified groups and generates a personalized search dictionary in accordance with the assigned groups and search dictionaries for the target user.

3. The system as in claim 2, wherein the personalized search dictionary is configured to be installed and operate in a mobile device.

4. The system as in claim 2, wherein the search dictionary component generates the personalized search dictionary from analysis of collective search query information that includes data related to search queries submitted by a predetermined user population over a period of time.

5. The system as in claim 4, wherein the collective search query information includes data related to search queries submitted by the user over a period of time.

6. The system as in claim 4, wherein the target user is not a member of the predetermined user population.

7. The system as in claim 4, wherein the user profile creating component performs the analysis of collective search query information by identifying a plurality of previously submitted search queries of a population of search users, determining groups of the search users based on the previously submitted search queries, and generating a group search dictionary for each one of the determined groups.

8. The system as in claim 7, wherein the target user is placed into one or more of the determined groups, and the personalized search dictionary is generated according to the user membership to each determined group into which the target user is placed.

9. The system as in claim 2, wherein the analysis of collective search query information is repeated at predetermined intervals on updated collective search query information.

10. The system as in claim 2, wherein the user profile creating component performs the analysis of collective search query information at predetermined intervals on updated collective search query information.

11. The system as in claim 2, wherein the query processing and user token profile component excludes a portion of the collective search query information from the query information processing.

12. The system as in claim 2, wherein the query processing and user token profile component performs tokenizing by identifying trigger pairs in the user profiles.

13. The system as in claim 2, wherein the personalized search dictionary component determines the personalized search dictionary that the target user should receive.

14. The system as in claim 2, wherein the personalized search dictionary is generated from among a predetermined set of group search dictionaries.

28. The system as in claim 1, wherein the search dictionary component further generates a personalized search dictionary, wherein the personalized search dictionary comprises at least one group dictionary selected from among a plurality of group dictionaries, each of which is associated with a user group and is produced from search query information from a user population comprising multiple users, wherein the personalized search dictionary includes a group dictionary associated with each user group with which the user is identified.

15. A method of generating a personalized search dictionary at a search service computer for query completion suggestion at a computing device, the method comprising:

creating a user profile at the search service computer for each user of a user population, each user profile identifying a user and corresponding raw search queries submitted by the identified user over a period of time;

tokenizing the user profiles, performing query information processing, including spelling correction, query tokenizing, and creating user-token profiles at the search service computer, in which search query terms of the raw search queries in each of the user profiles are parsed into tokens comprising letters and identifiable words and a user-token profile corresponding to each user profile is created comprising the raw search queries replaced with the tokenized search queries;

defining one or more groups based on the user-token profiles wherein each user of the user population is assigned to one or more of the defined groups, wherein each defined group comprises a collection of users that are indicated by frequency of the tokenized search queries as having similar preferences;

creating a group-specific search dictionary at the search service computer for each defined group in accordance with the tokenized search queries of the users in the corresponding defined group;

generating a personalized search dictionary for the user in accordance with data submitted by a predetermined user population over a period of time, wherein the personalized search dictionary provides the user with query completion suggestions that are configured to be submitted to a search facility in response to being accepted by the user.

16. The method as in claim 15, further including:

assigning a target user to one or more of the defined groups;

providing the target user with a personalized search dictionary in accordance with the one or more defined groups to which the target user is assigned.

17. The method as in claim 15, wherein the target user is a member of the user population.

18. The method as in claim 15, wherein a portion of the collective search query information is excluded from the tokenizing.

19. The method as in claim 15, wherein tokenizing includes identifying trigger pairs in the user profiles.

20. The method as in claim 15, wherein the personalized search dictionary is generated from analysis of collective search query information that includes data related to search queries submitted by a predetermined user population over a period of time.

21. The method as in claim 20, wherein the collective search query information includes data related to search queries submitted by the target user over a period of time.

22. The method as in claim 20, wherein the target user is not a member of the predetermined user population.

23. The method as in claim 20, wherein the analysis of collective search query information comprises:

identifying a plurality of previously submitted search queries of a population of search users;

determining groups of the search users based on the previously submitted search queries;

generating a group search dictionary for each one of the determined groups.

24. The method as in claim 23, wherein the target user is placed into one or more of the determined groups, and the personalized search dictionary is generated according to the user membership to each determined group into which the target user is placed.

25. The method as in claim 24, wherein the personalized search dictionary is configured to be installed and operate in a mobile device.

26. The method as in claim 20, wherein the analysis of collective search query information is repeated at predetermined intervals on updated collective search query information.

27. The method as in claim 15, further comprising:

generating a personalized search dictionary, wherein the personalized search dictionary comprises at least one group dictionary selected from among a plurality of group dictionaries, each of which is associated with a user group and is produced from search query information from a user population comprising multiple users, wherein the personalized search dictionary includes a group dictionary associated with each user group with which the user is identified.