PROVIDING MULTI-LINGUAL SEARCHING OF MONO-LINGUAL CONTENT

• US 20150058312A1
• Filed: 10/30/2014
• Published: 02/26/2015
• Est. Priority Date: 07/06/2012
• Status: Active Grant

• Current Assignee: Kyndryl Incorporated
• Original Assignee: International Business Machines Corporation
• Inventors: CASKEY, Sasha P., HAMILTON, Rick A. II, KANEVSKY, Dimitri, SAINATH, Tara N.
• Granted Patent: US 9,418,158 B2
• Time in Patent Office: Days
• US Class Current: 707/706
• CPC Class Codes: G06F 16/3337 Translation of the query language, e.g. Chinese to English
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/953 Querying, e.g. by the use of web search engines
• CPC Class Codes: G06F 40/129 Handling non-Latin characters, e.g. kana-to-kanji conversion
• CPC Class Codes: G06F 40/263 Language identification
• 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

Approaches for translating a transliterated search query are provided. An approach includes receiving a search query containing a transliterated word. The approach also includes determining a source language corresponding to the transliterated word. The approach further includes converting the transliterated word to a word in the source language. The approach additionally includes translating the word in the source language to a word in a target language. The approach also includes performing a search using the word in the target language.

Claims

1. A method implemented in a computer infrastructure, comprising:

receiving a search query containing a transliterated word;

determining a source language corresponding to the transliterated word, wherein the determining the source language comprises;

determining a weighted score for each one of a plurality of candidate languages, and designating the candidate language with the highest weighted score as the source language;

converting the transliterated word to a word in the source language;

translating the word in the source language to a word in a target language; and

performing a search using the word in the target language.

2. The method of claim 1, wherein the performing the search comprises submitting the word in the target language to an Internet search engine.

3. The method of claim 2, wherein the Internet search engine is indexed in the target language.

4. The method of claim 2, further comprising receiving search results from the Internet search engine based on the performing the search using the word in the target language.

5. The method of claim 4, wherein the search results are in the target language.

6. The method of claim 5, further comprising translating the search results, received from the Internet search engine, from the target language to the source language.

7. The method of claim 6, wherein the search query is received from a user computer device, and further comprising transmitting the translated search results in the source language to the user computer device.

8. The method of claim 1, wherein the determining the source language is performed automatically without a user specifying the source language.

9. The method of claim 1, wherein the converting the transliterated word to the word in the source language is based on visually analogous transliteration.

10. The method of claim 9, wherein the visually analogous transliteration comprises comparing a shape of a letter of the transliterated word to a shape of a letter in the source language.

11. The method of claim 1, wherein the converting the transliterated word to the word in the source language is based on a combination of visually analogous transliteration and phonetically analogous transliteration.

12. The method of claim 1, wherein the search query contains the transliterated word and a non-transliterated word.

13. The method of claim 1, wherein the transliterated word comprises a first transliterated word and the search query contains a second transliterated word, and further comprising:

determining a second source language corresponding to the second transliterated word;

converting the second transliterated word to a word in the second source language;

translating the word in the second source language to a second word in the target language; and

performing the search using the word in the target language and the second word in the target language.

14. A system implemented in hardware, comprising:

a computer system comprising a processor executing a translation engine that is configured to;

receive a search query containing a transliterated word;

determine a source language corresponding to the transliterated word, wherein the determining the source language comprises;

determining a weighted score for each one of a plurality of candidate languages, and designating the candidate language with the highest weighted score as the source language;

convert the transliterated word to a word in the source language;

translate the word in the source language to a word in a target language;

submit the word in the target language to an Internet search engine that is indexed in the target language;

receive search results in the target language, wherein the search results are based on the submitting the word in the target language to the Internet search engine; and

translate the search results, received from the Internet search engine, in the target language to search results in the source language.

15. The system of claim 14, wherein the converting the transliterated word to the word in the source language is based on visually analogous transliteration that comprises comparing a shape of a letter in the transliterated word to shapes of letters in the determined source language.

16. A method of deploying a system for translating a transliterated search query, comprising:

receive a search query from a user computer device, wherein the search query contains a transliterated word;

determine a source language corresponding to the transliterated word by determining a weighted score for each one of a plurality of candidate languages;

convert the transliterated word to a word in the source language;

translate the word in the source language to a word in a target language;

submit the word in the target language to a search engine that is indexed in the target language;

receive search results in the target language, wherein the search results are based on the submitting the word in the target language to a search engine;

translate the search results in the target language to search results in the source language; and

transmit the search results in the source language to the user computer device.

17. The method of claim 16, wherein the weighted score for each one of the plurality of candidate languages is based on at least one of:

correlating user-entered words against language-specific dictionaries;

environmental variables;

location-based services; and

user profile and/or history.

18. The method of claim 16, wherein the converting the transliterated word to the word in the source language is based on visually analogous transliteration.

19. The method of claim 16, wherein the weighted score for each one of the plurality of candidate languages is based on environmental variables that include data a browser.