Display of results of cross language search

• US 7,783,633 B2
• Filed: 11/19/2004
• Issued: 08/24/2010
• Est. Priority Date: 11/19/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Franz, Martin, Ward, Robert Todd, McCarley, Jeffrey Scott, Florian, Radu
• Primary Examiner(s): Vo; Tim T.
• Assistant Examiner(s): SMITH, GARRETT A
• Application Number:
  US10/992,919
• Publication Number:
  US 20060111893A1
• Time in Patent Office: 2,104 Days
• Field of Search: 715/247, 715/256, 715/264
• US Class Current: 707/729
• CPC Class Codes: G06F 40/58 Use of machine translation, e.g. for multi-lingual retrieval, for server-side translation for client devices or for real-time translation

Abstract

An improved method, apparatus, and computer instructions displaying search results. A query is received in a first language. A snippet in the document containing the word is selected in response to identifying a document in a second language in which the document contains a word matching a query word in the query. This snippet is translated into the first language.

Claims

1. A method in a data processing system for displaying a search result, the method comprising:

receiving a query in a first language;

translating, by the data processing system, the query into a second language to form a token;

identifying a target document, wherein the target document is identified by performing a search based on the token;

scanning the target document for the token;

responsive to scanning the target document, selecting, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language;

responsive to selecting the snippet, translating, automatically, the snippet into the first language; and

storing the snippet;

wherein the predefined set of rules comprises a rule to select a boundary for the snippet using a combination of the token and a predefined number of adjacent words and wherein the boundary is selected to keep phrases and clauses unbroken.

2. The method of claim 1 further comprising:

displaying the snippet as the search result.

3. The method of claim 1 further comprising:

selecting a second snippet from the target document, wherein the second snippet is based on an additional instance of the token.

4. The method of claim 3 further comprising:

translating the second snippet into the first language.

5. The method of claim 3, wherein the second snippet is selected to maximize translation scores of document language words.

6. The method of claim 1 wherein the token comprises a first word and a second word and wherein the method further comprises:

selecting the snippet based on the first word; and

selecting a second snippet based on the second word.

10. The method of claim 1, wherein the step of responsive to scanning the target document, selecting, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language, further comprises:

calculating a score for each snippet of a plurality of snippets using the formula

$\underset{i}{\sum <br><br>}<br><br>$

where q_i and d_j run over the query words and the document words, respectively, and t(q_i|q_i) is an estimate of the probability of translating the word d_j into q_i;

identifying a highest score; and

selecting the snippet based on the highest score.

7. A method in a data processing system for displaying a search result, the method comprising:

receiving a query in a first language;

translating, by the data processing system, the query into a second language to form a token;

identifying a target document, wherein the target document is identified by performing a search based on the token;

scanning the target document for the token; and

displaying a graphical indicator, wherein the graphical indicator indicates a presence of at least one instance of the token in the target document and a relative position of the at least one token within the target document;

responsive to scanning the target document, selecting, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language;

responsive to selecting the snippet, translating, automatically, the snippet into the first language to form a translated snippet; and

displaying the translated snippet as the search result;

wherein the predefined set of rules comprises a rule to select a boundary for the snippet using a combination of the token and a predefined number of adjacent words and wherein the boundary is selected to keep phrases and clauses unbroken.

8. The method of claim 7, wherein the graphical object is a matching bar.

9. The method of claim 8, wherein the graphical indicator is a highlight.

11. The method of claim 7, wherein the step of responsive to scanning the target document, selecting, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language, further comprises:

calculating a score for each snippet of a plurality of snippets using the formula

$\underset{i}{\sum <br><br>}<br><br>$

where q_i and d_j run over the query words and the document words, respectively, and t(q_i|q_i) is an estimate of the probability of translating the word d_j into q_i;

identifying a highest score; and

selecting the snippet based on the highest score.

12. A computer program product comprising:

computer recordable medium having a plurality of computer useable instructions stored thereon for execution by a computer, the plurality of computer useable instructions adopted to cause a processor of the computer to perform steps comprising;

receiving a query in a first language;

translating the query into a second language to form a token;

identifying a target document, wherein the target document is identified by performing a search based on the token;

scanning the target document for the token;

responsive to scanning the target document, selecting, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language;

responsive to selecting the snippet, translating, automatically, the snippet into the first language; and

storing the snippet;

wherein the predefined set of rules comprises a rule to select a boundary for the snippet using a combination of the token and a predefined number of adjacent words and wherein the boundary is selected to keep phrases and clauses unbroken.

13. The computer program product of claim 12, wherein the plurality of computer useable instructions are further adapted to select the snippet by performing steps comprising:

calculating a score for each snippet of a plurality of snippets using the formula

$\underset{i}{\sum <br><br>}<br><br>$

where q_i and d_j run over the query words and the document words, respectively, and t(q_i|q_i) is an estimate of the probability of translating the word d_j into q_i;

identifying a highest score; and

selecting the snippet based on the highest score.

14. A data processing system comprising:

a bus;

a storage device connected to the bus, wherein the storage device contains a plurality of computer usable instructions; and

a processing unit connected to the bus, wherein the processing unit executes the plurality of computer usable instructions to receive a query in a first language;

translate the query into a second language to form a token;

identify a target document, wherein the target document is identified by performing a search based on the token;

scan the target document for the token;

responsive to scanning the target document, select, automatically, based on a set of predefined rules, a snippet associated with at least one instance of the token in the target document, wherein the snippet is less than an entirety of the target document, and wherein the snippet is in the second language;

responsive to selecting the snippet, translate, automatically, the snippet into the first language; and

store the snippet;

wherein the predefined set of rules comprises a rule to select a boundary for the snippet using a combination of the token and a predefined number of adjacent words and wherein the boundary is selected to keep phrases and clauses unbroken.

15. The apparatus of claim 14, wherein the processing unit further executes the plurality of computer useable instructions to select the snippet by:

executing the computer useable instructions to calculate a score for each snippet of a plurality of snippets using the formula

$\underset{i}{\sum <br><br>}<br><br>$

where q_i and d_j run over the query words and the document words, respectively, and t(q_i|q_i) is an estimate of the probability of translating the word d_j into q_i; and

select a snippet with the highest score;

identify a highest score; and

select the snippet based on the highest score.