Search keyword improvement apparatus, server and method

• US 9,146,999 B2
• Filed: 02/17/2009
• Issued: 09/29/2015
• Est. Priority Date: 03/27/2008
• Status: Expired due to Fees

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Suzuki, Masaru, Kokubu, Tomoharu
• Primary Examiner(s): Kim, Taelor
• Application Number:
  US12/372,570
• Publication Number:
  US 20090248674A1
• Time in Patent Office: 2,415 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A search keyword improvement apparatus includes a unit extracting a word as an additional keyword candidate from a new document, number of times of appearance of the word in the new document being greater than number of times of appearance of the word in each of a first documents except for the new document, if the new document and a new search target identification information item which is used to search the new document are accumulated, a unit generating a first search query based on an input keyword, a second search target associated with the input keyword, and one of the additional keywords, and generating a second search query, a unit moving the additional keyword candidate and the third search target identification information item, if the desired search result is selected from a third search result list corresponding to the second search query.

Claims

1. A search keyword improvement apparatus comprising:

a computer;

a history accumulation unit resident on the computer to accumulate first documents or indexes which correspond to the first documents and first search target identification information items which correspond to first search targets associated with keywords for searching, the first documents or the indexes and the first search target identification information items being in relationship to each other on the history accumulation unit, the first search target identification information items being used to search the first documents or indexes;

an additional keyword candidate extraction unit resident on the computer to extract a word, as a first additional keyword candidate from a new document included in the first documents or indexes;

if the new document and a new search target identification information item, which is used to search the new document, are accumulated in the history accumulation unit, then the additional keyword candidate extraction unit extracts the first additional keyword candidate when a computed statistical score for the first additional keyword candidate is greater than a preset threshold value so that a number of times of appearance of the first additional keyword candidate in the new document is greater than number of times of appearance of the first additional keyword candidate in each of the first documents or indexes other than new document;

an additional keyword accumulation unit resident on the computer to accumulate additional keywords different from the first additional keyword candidate and second search target identification information items corresponding to the additional keywords;

an additional keyword candidate holding unit resident on the computer to store additional keyword candidates including the first additional keyword candidate and different from the additional keywords and a third search target identification information item corresponding to the first additional keyword candidate;

a search query generation unit resident on the computer to generate a first search query based on an input keyword, a second search target associated with the input keyword, and one of the additional keywords which corresponds to a fourth search target identification information item corresponding to the second search target, and configured to generate a second search query based on the input keyword, the second search target, and the first additional keyword candidate which corresponds to the fourth search target identification information item;

a search result list generating unit resident on the computer to generate a first search result list including search results corresponding to the first search query and the second search query;

a first selection unit resident on the computer to select a second document as a desired search result from the first search result list; and

an additional keyword updating unit resident on the computer to move, to the additional keyword accumulation unit, the first additional keyword candidate and the third search target identification information item, if the first selection unit selects the desired search result from a second search result list corresponding to the second search query, the second search result list being included in the first search result list.

2. The apparatus according to claim 1, wherein the first selection unit selects the second document from the first search result list if a user has determined that the second document is the desired search result.

3. The apparatus according to claim 2, wherein the second search target in the search query generation unit is a user selected one of the first search targets.

4. The apparatus according to claim 1, further comprising:

a keyword input unit configured to input the input keyword;

a keyword analysis unit configured to analyze a semantic class of the input keyword; and

a search target presentation unit configured to display a search target menu including the second search target in accordance with an analysis result obtained by the keyword analysis unit,

the second search target associated with the input keyword being selected from the search target menu.

5. The apparatus according to claim 1, wherein the additional keyword candidate extraction unit extracts the word as the first additional keyword candidate if a value of (n/N)/(m/M) is greater than a first threshold value, n indicating the number of times of appearance of the word in the new document, N indicating number of times of appearance of all words in the new document, m indicating number of times of appearance of the word in the first documents or indexes except for the new document, M indicating number of times of appearance of all words in the first documents or indexes except for the new document.

6. The apparatus according to claim 1, wherein the search result list generation unit includes:

a determination unit configured to determine whether there is a first search result corresponding to the second search query;

a first read unit configured to read the first search result if it is determined that there is the first search result;

a second read unit configured to read a second search result corresponding to the first search query; and

a deletion unit configured to delete, from the first search result, a document included in the second search result.

7. The apparatus according to claim 6, wherein the search result list generation unit further includes:

a reset unit configured to reset content of the first search result list to obtain a reset list; and

an update unit configured to update the first search result list by performing an operation of moving, to the reset list, a first result element of a highest rank included in the second search result, and then moving, to a position immediately after the first result element of the reset list, a second result element of a highest rank included in the first search result, and repeating the operation until all search result elements included in the first search result and the second search result are processed.

8. The apparatus according to claim 1, wherein the first selection unit includes:

a measurement unit configured to measure a time for which a third document included in the second search result list and browsed by a user is continuously displayed; and

a second selection unit configured to select the third document as the desired search result if the measured time is greater than a second threshold value.

9. The apparatus according to claim 8, wherein the first selection unit further includes:

a determination unit configured to determine whether the third document is a search result of the second search query; and

an identification unit configured to identify a fifth search target identification information item corresponding to the third document if the determination unit determines that the third document is the search result of the second search query.

10. The apparatus according to claim 1, wherein the first search result list generated by the search result list generation unit includes, in units of search results, information corresponding to one of the first search query and the second search query.

11. A server apparatus comprising:

a server;

a history accumulation unit resident on the server to accumulate first documents selected as desired search results from a search result list formed of search results, and search target identification information items which correspond to search targets associated with keywords for searching, the first documents and the search target identification information items being in relationship to each other on the history accumulation unit, the search target identification information items being used to search the first documents;

an additional keyword candidate extraction unit resident on the server to extract a word, as first additional keyword candidate from a new document included in the first documents;

if the new document and a new search target identification information item, which is used to search the new document, are accumulated in the history accumulation unit, then the additional keyword candidate extraction unit extracts the first additional keyword candidate when a computed statistical score for the first additional keyword candidate is greater than a preset threshold value so that a number of times of appearance of the first additional keyword candidate in the new document is greater than number of times of appearance of the first additional keyword candidate in each of the first documents other than new document; and

an additional keyword candidate holding unit resident on the server to store the first additional keyword candidate and another search target identification information item corresponding to the first additional keyword candidate.

12. A search keyword improvement method comprising:

preparing a history accumulation unit configured to accumulate first documents or indexes which correspond to the first documents and first search target identification information items which correspond to first search targets associated with keywords for searching, the first documents or the indexes and the first search target identification information items being in relationship to each other on the history accumulation unit, the first search target identification information items being used to search the first documents or indexes;

extracting a word as first additional keyword candidate from a new document included in the first documents or indexes;

if the new document and a new search target identification information item, which is used to search the new document, are accumulated in the history accumulation unit, then extracting the first additional key word candidate when a computed statistical score for the first additional keyword candidate is greater than a preset threshold value so that a number of times of appearance of the first additional keyword candidate in the new document is greater than a number of times of appearance of the first additional keyword candidate in each of the first documents or indexes other than the new document;

preparing an additional keyword accumulation unit configured to accumulate additional keywords different from the first additional keyword candidate and second search target identification information items corresponding to the additional keywords;

storing in a storage unit additional keyword candidates including the first additional keyword candidate and different from the additional keywords and a third search target identification information item corresponding to the first additional keyword candidate;

generating a first search query based on an input keyword, a second search target associated with the input keyword, and one of the additional keywords which corresponds to a fourth search target identification information item corresponding to the second search target;

generating a second search query based on the input keyword, the second search target, and the first additional keyword candidate which corresponds to the fourth search target identification information item;

generating a first search result list including search results corresponding to the first query and the second query;

selecting a second document as a desired search result from the first search result list; and

moving, to the additional keyword accumulation unit, the first additional keyword candidate and the third search target identification information item, if the first selection unit selects the desired search result from a second search result list corresponding to the second search query, the second search result list being included in the first search result list.

13. A method comprising:

preparing a history accumulation unit configured to accumulate first documents selected as desired search results from a search result list formed of search results, and search target identification information items which correspond to search targets associated with keywords for searching, the first documents and the search target identification information items being in relationship to each other on the history accumulation unit, the search target identification information items being used to search the first documents;

extracting a word, as first additional keyword candidate from a new document included in the first documents;

if the new document and a new search target identification information item, which is used to search the new document, are accumulated in the history accumulation unit, then extracting the first additional key word candidate when a computed statistical score for the first additional keyword candidate is greater than a preset threshold value so that a number of times of appearance of the first additional keyword candidate in the new document is greater than a number of times of appearance of the first additional keyword candidate in each of the first documents other than the new document; and

storing in a storage unit the additional keyword candidates including the first additional keyword candidate and different from the additional keywords and another search target identification information item corresponding to the first additional keyword candidate.