User-context-based search engine

• US 20070033218A1
• Filed: 08/08/2006
• Published: 02/08/2007
• Est. Priority Date: 08/08/2005
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Taylor, David
• Granted Patent: US 8,429,167 B2
• Time in Patent Office: Days
• 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 creating a universal vocabulary, the method comprising:

selecting a finite number of domains;

creating domain lists for each domain, the domain lists each comprising selected terms substantially unique to the associated domain; and

building a vocabulary list by identifying a corpus of information organized by topical entries, counting occurrences of terms from the domain list for each topical entry, and calculating a macro-context for each topical entry by generating a vector associating occurrences of domain list terms with the corresponding domain.

2. 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 associated with a macro-context;

counting occurrences of each term from the vocabulary list found;

calculating a macro-context representing summations of the macro-contexts associated with the terms from the vocabulary list found within 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.

3. The method of claim 2, 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.

4. A method for searching comprising:

mining a repository or information to determine macro and micro-contexts for elements 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;

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

presenting the information located to a user.

5. The method of claim 4, wherein the repository comprises a database.

6. The method of claim 4, wherein the macro-context is a vector representing weights corresponding to the use of vocabulary terms from a list of topical entries.

7. The method of claim 4, wherein the micro-contexts represent a list of words most closely reflecting the content of the contexts of information in the repository.

8. The method of claim 4, 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.

9. The method of claim 4, 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.

10. The method of claim 4, 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.

11. The method of claim 10, wherein the context compared is selected from macro-context and micro-context.