Protecting information in search queries

• US 9,195,744 B2
• Filed: 07/20/2009
• Issued: 11/24/2015
• Est. Priority Date: 07/25/2008
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adachi, Noritaka, Kato, Shingo, Kawanaka, Shinya, Rudy, Raymond Harry Putra
• Primary Examiner(s): HUANG, MIRANDA M
• Application Number:
  US12/506,109
• Publication Number:
  US 20100023509A1
• Time in Patent Office: 2,318 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24534 Query rewriting; Transformation
• CPC Class Codes: G06F 16/24535 of sub-queries or views
• CPC Class Codes: G06F 16/24564 Applying rules; Deductive queries
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques

Abstract

A method of protecting information in search queries uses a search apparatus with a user interface that is configured for connection to a computer network that comprises a plurality of search engines on a plurality of servers. The method includes receiving a search query comprising a plurality of keywords; dividing the search query into a number of sub-queries, each sub-query comprising at least one of the keywords; and submitting the sub-queries to different search engines.

Claims

1. A search method, said method comprising:

accepting user input of a main query including a plurality of keywords;

accepting user input that identifies at least one combination of said keywords inhibited to be used as a sub-query;

generating a plurality of sub queries based on the user input each including one or more keywords by combining the keywords included in the input main query;

excluding sub-queries that are identified as inhibited combinations;

calculating a co-occurrence probability for the keywords of each sub-query;

sorting keyword combinations in a descending order based on the co-occurrence probabilities;

excluding from use as a sub-query any combination of keywords having a co-occurrence probability below a threshold;

separately sending the generated sub queries to a number of search engines;

re-searching a plurality of search results returned from the search engine in response to the plurality of sub queries, with the input main query; and

displaying search results acquired by the re-searching step.

2. The method of claim 1, further comprising searching said keywords in a dictionary and excluding from use as a sub-query any combination of said keywords containing a keyword not found in said dictionary.

3. The method of claim 1, further comprising calculating a co-occurrence probability for the keywords of each sub-query and excluding from use as a sub-query a specific number of keyword combinations having the lowest co-occurrence probability.

4. The method of claim 1, further comprising receiving individual search results from each of said search engines, each individual search result comprising a number of hits matching the sub-query submitted to a corresponding search engine.

5. The method of claim 4, in which said individual search results each comprise a listing of web pages.

6. The method of claim 4, further comprising filtering said individual search results to exclude listings that do not satisfy said search query.

7. The method of claim 6, further comprising integrating said individual search results from said search engines to produce an integrated search result.

8. A search apparatus implemented in a computer, said apparatus comprising a processor programmed to:

receive a search query comprising a plurality of keywords;

determine each mathematical combination of the plurality of keywords;

divide the search query into a number of sub-queries, in which each sub-query comprising at least one of said keywords and in which each mathematical combination is represented as a separate sub-query;

compare each sub-query to an inhibited combinations list;

exclude sub-queries that are identified as inhibited combinations;

after excluding sub-queries that are identified as inhibited combinations, submit remaining sub-queries to different search engines such that private information from the search query is less discernible at any one of said search engines than if more of the keywords of the search query were provided to that individual search engine; and

re-search a plurality of search results returned from the different search engines in response to the submission of said sub-queries, said re-searching being performed using all the keywords of the search query.

9. The search apparatus of claim 8, in which said processor is further programmed to accept user input identifying at least one keyword or combination of said keywords that is not to be used as a sub-query.

10. The search apparatus of claim 8, in which said processor is further programmed to search said keywords in a dictionary and to exclude from use as a sub-query any combination of said keywords containing a keyword not found in said dictionary.

11. The search apparatus of claim 8, in which said processor is further programmed to calculate a co-occurrence probability for the keywords of each sub-query and to exclude from use as a sub-query any combination of keywords having a co-occurrence probability below a threshold, in which the co-occurrence probability is a ratio of percentage chance that multiple keywords will be used in a same search result.

12. The search apparatus of claim 8, in which said processor is further programmed to calculate a co-occurrence probability for the keywords of each sub-query and to exclude from use as a sub-query a specific number of keyword combinations having the lowest co-occurrence probability.

13. The search apparatus of claim 8, in which said processor is further programmed to receive individual search results from each of said search engines, each individual search result comprising a number of hits matching the sub-query submitted to a corresponding search engine.

14. The search apparatus of claim 13, in which said processor is further programmed to filter said individual search results to exclude listings that do not satisfy said search query.

15. The search apparatus of claim 14, in which said processor is further programmed to integrate search results from each of said search engines to produce an integrated search result.

16. A computer program product for protecting information in search queries, the computer program product comprising:

a tangible computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising;

computer readable program code configured to receive a search query comprising a plurality of keywords;

computer readable program code configured to determine each mathematical combination of the plurality of keywords;

computer readable program code configured to divide the search query into a number of sub-queries, in which each sub-query comprising at least one, but less than all, of said keywords and in which each mathematical combination is represented as a separate sub-query;

computer readable program code configured to exclude sub-queries identified as inhibited combinations and based on a co-occurrence probability, in which the co-occurrence probability is a ratio of percentage chance that multiple keywords will be used in a same search result;

computer readable program code configured to submit remaining sub-queries to different search engines after said exclusion such that each search engine receives only a sub-query and less than all of said plurality of keywords such that private information from the search query is less discernible at any one of said search engines than if that search engine received more of said plurality of keywords of said search query;

computer readable program code configured to, generate an integrated search results based on a rank of hits returned from the different search engines in response to the submission of said sub-queries, in which hits that share rank are sorted based on a number of rules, in which the number of rules comprise;

a first rule indicating a hit returned from a priority search engine is ranked above a hit from another search engine;

a second rule indicating a hit returned by more search engines is ranked above a hit returned by fewer search engines; and

a third rule indicating a hit returned from a search engine with more total number of hits is ranked above a hit returned from a search engine with fewer total number of hits;

in which the first rule is evaluated before the second rule and in which the second rule is evaluated before the third rule;

computer readable program code configured to search the integrated search results.

17. The computer program product of claim 16, further comprising computer readable code configured to exclude specific keywords or combinations of keywords from use as a sub-query.

18. The computer program product of claim 16, further comprising computer readable code configured to:

filter individual search results to exclude listings that do not satisfy said search query.

19. A method of protecting information in search queries using a search apparatus with a user interface that is configured for connection to a computer network, said network comprising a plurality of search engines on a plurality of servers, said method comprising:

receiving a search query comprising a plurality of keywords;

dividing the search query into a number of sub-queries, each sub-query comprising at least one of said keywords but less than all of said keywords;

comparing each sub-query to an inhibited combinations list that includes a number of keywords from the search query;

excluding sub-queries that are identified as inhibited combinations;

submitting remaining sub-queries to different search engines such that each search engine receives less than all of said plurality of keywords such that private information from the search query is less discernible at any one of said search engines than if that search engine received more of said plurality of keywords of said search query; and

generating an integrated search results based on a rank of hits included in individual search results returned from the different search engines in response to the submission of said sub-queries, in which hits that share rank are sorted based on a number of priority rules; and

searching the integrated search results.

20. The method of claim 19, further comprising accepting user input that identifies at least one keyword or combination of said keywords that is not to be used as a sub-query.

21. The method of claim 19, further comprising searching said keywords in a dictionary of the language in which said search query is presented and excluding from use as a sub-query any combination of said keywords containing a keyword not found in said dictionary.

22. The method of claim 19, further comprising calculating a co-occurrence probability for the keywords of each sub-query and excluding from use as a sub-query any combination of keywords having a co-occurrence probability below a threshold.

23. The method of claim 19, further comprising calculating a co-occurrence probability for the keywords of each sub-query and excluding from use as a sub-query a specific number of keyword combinations having the lowest co-occurrence probability.