User-context-based search engine

• US 9,449,105 B1
• Filed: 09/14/2012
• Issued: 09/20/2016
• Est. Priority Date: 08/08/2005
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Taylor, David C.
• Primary Examiner(s): Shanmugasundaram, Kannan
• Application Number:
  US13/619,839
• Time in Patent Office: 1,467 Days
• Field of Search: 707/708, 707/739, 707/794
• US Class Current: 1/1
• CPC Class Codes: G06F 16/374 Thesaurus
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

A method and apparatus for determining contexts of information analyzed. Contexts may be determined for words, expressions, and other combinations of words in bodies of knowledge such as encyclopedias. Analysis of use provides a division of the universe of communication or information into domains, and selects words or expressions unique to those domains of subject matter as an aid in classifying information. A vocabulary list is created with a macro-context (context vector) for each, dependent upon the number of occurrences of unique terms from a domain, over each of the domains. This system may be used to find information or classify information by subsequent inputs of text, in calculation of macro-contexts, with ultimate determination of lists of micro-contests including terms closely aligned with the subject matter.

Claims

1. A method for classifying information, the method comprising:

providing input text;

identifying a vocabulary list independent from the input text, the vocabulary list comprising a plurality of entries, each of the plurality of entries associated with a macro-context, wherein the macro-context comprises a vector characterizing the context of the entry by mapping a plurality of subject matters, each unique, to a corresponding plurality of weights, each weight reflecting a contribution of a corresponding subject matter of the plurality of subject matters to the entry;

counting occurrences of each term from the vocabulary list found within the input text;

calculating a macro-context representing summations of the macro-contexts associated with the terms from the vocabulary list found within the input text to characterize the context of the input text by mapping the plurality of subject matters to corresponding weights reflecting contributions of corresponding subject matters of the plurality of subject matters to the input text; and

determining a micro-context comprising a list of terms selected from the list of vocabulary that correspond to the input text.

2. The method of claim 1, further comprising reducing the list of terms in the micro-context by selecting those terms having macro-contexts most closely aligned with the macro-context of the input text.

3. The method of claim 1, wherein determining a micro-context comprises identifying, from the plurality of entries, entries having a macro-context within a selected mathematical proximity to the calculated macro-context.

4. The method of claim 1, wherein the plurality of entries comprises a plurality of topical entries, each entry of the plurality of entries corresponding to a topical entry of the plurality of topical entries.

5. The method of claim 1, wherein the macro-context of each of the plurality of entries reflects the counted occurrences.

6. The method of claim 1, wherein determining a micro-context further comprises calculating a multiplication of at least two macro-context vectors to provide a mathematical value reflecting a correspondence of a query with a vocabulary list of topical entries.

7. The method of claim 1, wherein each of the plurality of entries comprises at least one of a word, a name, or a phrase.

8. The method of claim 1, providing input text comprising at least one of providing a body of a web page to be characterized, providing a query from a user, or providing a query history of a user, the query history including previous queries submitted by the user and responses to the previous queries of the user.

9. The method of claim 1, further comprising classifying the input text according to at least one of the macro-context or the micro-context.

10. A method for searching comprising:

mining a repository of information to determine macro and micro-contexts for elements of a database, each macro and micro-contexts characterizing the context of an element of the database by mapping a plurality of subject matters, each unique, to a corresponding plurality of weights, each weight reflecting a contribution of a corresponding subject matter of the plurality of subject matters to the element of the database;

indexing the database content according to the macro and micro-contexts determined;

receiving a query from a user;

determining macro and micro-contexts associated with the query, the macro and micro-contexts characterizing the context of the query by mapping the plurality of subject matters to corresponding weights reflecting contributions of corresponding subject matters of the plurality of subject matters to the query;

locating in a database information having contexts related to contexts associated with a query; and

presenting the information located to a user.

11. The method of claim 10, wherein the repository comprises the database.

12. The method of claim 10, wherein the macro-context is a vector representing weights corresponding to the use of vocabulary terms from a list of topical entries.

13. The method of claim 10, wherein the micro-contexts represent a list of words most closely reflecting the content of the contexts of information in the repository.

14. The method of claim 10, wherein receiving a query from a user further comprises receiving additional information from a user selected from the group consisting of previous queries by the user, previous responses to previous queries from a user, previous results from browsing by a user, and documents provided by a user to establish contexts.

15. The method of claim 10, wherein providing a micro-context further comprises calculating a multiplication of at least two macro-context vectors to provide a mathematical value reflecting a correspondence of a query with a vocabulary list of topical entries.

16. The method of claim 10, wherein locating further comprises comparing a parameter reflecting at least one context associated with the query to a corresponding parameter corresponding to a context of selected information from the repository of information.

17. The method of claim 16, wherein the context compared is selected from macro-context and micro-context.

18. The method of claim 10, wherein the database comprises a plurality of fields of text associated with at least one of a web page, a web site, or a group of web pages grouped on a web site under a heading.

19. The method of claim 10, wherein the repository of information comprises a plurality of terms and a plurality of topical entries, each term of the plurality of terms corresponding to a topical entry of the plurality of topical entries.

20. The method of claim 10, wherein locating in a database information comprises comparing and matching the macro and micro-contexts of the database with the macro and micro-contexts of the query.