EXPANSION OF SEARCH QUERIES USING INFORMATION CATEGORIZATION

• US 20110258173A1
• Filed: 06/28/2011
• Published: 10/20/2011
• Est. Priority Date: 12/04/2008
• Status: Active Grant

• Current Assignee: Alexander Rubinov, Dmitry Kuharenko, Michael Ratiner
• Original Assignee: Alexander Rubinov, Dmitry Kuharenko, Michael Ratiner
• Inventors: RUBINOV, Alexander, KUHARENKO, Dmitry, RATINER, Michael
• Granted Patent: US 8,219,579 B2
• Time in Patent Office: Days
• US Class Current: 707/706
• CPC Class Codes: G06F 16/332 Query formulation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A computerized system and method of constructing and expanding search queries for conducting searches through information sources. The system enables retrieving a category options tree, allowing a user to define a category route by selecting a category-node, which defines a search-category. The system may further enable retrieving a query scenario tree, having a hierarchal structure comprising query nodes, where the retrieved query scenario tree is associated with an initial input query, inputted by a user. Each query node defines a query route enabling to construct the content and structure of an expanded search query. The system enables selecting a query node of the retrieved query scenario tree, according to an online decision making process, which analyses the search-category in relation to available query routes in to allow selecting a query node from the retrieved scenario tree that is most compatible with the search-category.

Claims

1. A query builder operated by at least one processing unit for supporting query building processes, said query builder comprising:

a categorization module, which allows constructing a categorization scenario, which is a logic tree of search categories, wherein said categorization scenario represents statistical relations between categories, allowing the user to select a category therefrom; and

a query scenario module, which receives a selected category and dynamically builds a query scenario corresponding to said selected category, said query scenario is a logic tree having nodes of search terms structured in a hierarchal manner that represents interrelations between said search term nodes, said interrelations are determined according to said selected category and according to browsing statistics relating to searches done by users, wherein upon selection of a node therefrom a search query is automatically created, said search query comprises all search terms that are interrelated to said selected node according to the hierarchal structure of said query scenario.

2. The query builder according to claim 1 further comprising a graphical user interface (GUI), which allows presenting categories and query scenarios to users using end user devices and for allowing the users to select nodes and input data thereby.

3. The query builder according to claim 2, wherein said GUI further enables presenting the search query to the user and allowing the user to edit at least one of:

the search terms of the search query, the interrelations between the search terms.

4. The query builder according to claim 1, wherein said query scenario module determines said interrelations between nodes of search terms thereof by correlating said selected category with indexes of web pages and by using statistical browsing information, using at least one database indicative of browsing and indexing information relating to visited web pages.

5. The query builder according to claim 4, wherein the search terms in the query scenario are determined by using an algorithm that assigns a correlation level to each of said web pages'"'"' indexes, said correlation levels defines the correlation between the selected category and the index of each of the web pages indicated in said database, wherein said search terms are associated with web pages having a correlation level that is above a threshold value, and wherein the interrelations between the search terms defining the hierarchal structure of said query scenario are determined according to browsing statistics of the associated web pages.

6. The query builder according to claim 1, wherein said query scenario module further enables determining term-relations between all search terms of each created search query according to the interrelations between those terms as defined by the hierarchal structure of the respective query scenario.

7. The query builder according to claim 6, wherein said query scenario module further enables storing all search terms of each created search query and the term-relations therebetween for enhancing browsing statistics and using said saved search queries in determining structure of query scenarios.

8. The query builder according to claim 6 is operatively associated with at least one search engine, said search engine receives said search query indicative of term-relations between the search terms thereof from said query builder and uses said search query for searching through at least one communication network for information relating to the search query, using at least one searching technique.

9. The query builder according to claim 1 further comprising a language module for allowing presenting the user with equivalent query scenarios in other languages selected by the user, upon the building of an original query scenario at a first language, by retrieving or building an equivalent query scenario of said original first query scenario according to translations of search term of the original query scenario.

10. A computer implemented method of supporting search queries building processes, said method comprising:

providing a categorization scenario, which is a logic tree of search categories, said categorization scenario represents statistical relations between categories, using a query builder operated by at least one processing unit to create and present said categorization scenario;

manually selecting a category from said categories scenario by selecting a node therefrom, wherein a user selects said category;

dynamically building a query scenario corresponding to said selected category, said query scenario is a logic tree having nodes of search terms structured in a hierarchal manner that represents interrelations between said search term nodes, said interrelations are determined according to said selected category and according to browsing statistics, wherein said query builder builds and presents said query scenario;

manually selecting a node of said query scenario to define a search query thereby, wherein the user selects said category; and

automatically creating a search query comprising all search terms related to said node according to the hierarchal structure thereof, wherein said query builder creates said search query.

11. The method according to claim 10, wherein said interrelations between nodes of search terms are determined by correlating said selected category with indexes of web pages and by using statistical browsing information, using at least one database indicative of browsing and indexing information relating to visited web pages.

12. The method according to claim 11, wherein the search terms in the query scenario are determined by using an algorithm that assigns a correlation level to each of said web pages'"'"' indexes, said correlation levels defines the correlation between the selected category and the index of each of the web pages indicated in said database, wherein said search terms are associated with web pages having a correlation level that is above a threshold value, and wherein the interrelations between the search terms defining the hierarchal structure of said query scenario are determined according to browsing statistics of the associated web pages.

13. The method according to claim 10 further comprising determining term-relations between all search terms of each created search query according to the interrelations between those terms as defined by the hierarchal structure of the respective query scenario.

14. The method according to claim 13 further comprising storing all search terms of each created search query and the term-relations therebetween for enhancing browsing statistics and for using said saved search queries in determining structure of query scenarios.

15. The method according to claim 13 further comprising receiving said search query indicative of term-relations between the search terms thereof from said query builder and searching through at least one communication network for information relating to the search query, using said search query, using at least one searching technique.

16. The method according to claim 10 further comprising presenting the search query to the user and allowing the user to edit at least one of:

the search terms of the search query, the interrelations between the search terms.

17. A system for supporting query building processes, said system comprising:

at least one database comprising categories and information associated with interrelations between said categories for enabling to construct categories scenarios thereby; and

at least one processing unit that operates a query builder, which enables assisting a user in building a search query by constructing a categorization scenario using said database and presenting thereof to the user, said categories scenario is a logic tree of search categories, which represents statistical relations between categories and dynamically building a query scenario for each category defined by the user by selecting a category form said categories scenario, said query scenario is a logic tree having nodes of search terms structured in a hierarchal manner that represents interrelations between said search term nodes, said interrelations are determined according to said selected category and according to browsing statistics relating to searches done by users, wherein upon selection of a node therefrom a search query is automatically created, said search query comprises all search terms that are interrelated to said selected node according to the hierarchal structure of said query scenario.