Validating translations using visual similarity between visual media search results
First Claim
1. A computer-implemented method comprising:
- receiving a plurality of candidate translations for a phrase, wherein the phrase is in a first language and each candidate translation is a translation of the phrase into a different language from the first language;
obtaining a first group of visual media search results, wherein the first group of visual media search results are responsive to a first visual media search query with text corresponding to the phrase;
obtaining a respective second group of visual media search results for each of the candidate translations, wherein the second group of visual media search results are responsive to a second visual media search query with text corresponding to the candidate translation;
generating a respective quality of results statistic for each of the candidate translations, wherein the quality of results statistic for each of the candidate translations is a value that represents a quality of search results that are responsive to the second visual media search query with text corresponding to the candidate translation;
calculating, by one or more computers, a respective similarity score for each of the candidate translations, wherein the similarity score is an estimate of visual similarity between the first group of visual media search results and the second group of visual media search results for the candidate translation;
selecting one or more of the candidate translations based on, for each candidate translation, the similarity score for the candidate translation and the quality of results statistic for the candidate translation; and
associating each of the one or more selected candidate translations with the phrase as a valid translation for the phrase.
2 Assignments
0 Petitions
Accused Products
Abstract
Methods, systems, and apparatus, including computer program products, for validating candidate translations for a phrase. A plurality of candidate translations for a phrase are received. A similarity score for each of the candidate translations is calculated. Each similarity score is an estimate of visual similarity between a group of visual media search results responsive to a visual media search query with text corresponding to the phrase and a group of visual media search results responsive to a visual media search query with text corresponding to a candidate translation. One or more of the candidate translations are selected based on the similarity scores. The selected candidate translations are associated with the phrase as valid translations for the phrase.
104 Citations
24 Claims
-
1. A computer-implemented method comprising:
-
receiving a plurality of candidate translations for a phrase, wherein the phrase is in a first language and each candidate translation is a translation of the phrase into a different language from the first language; obtaining a first group of visual media search results, wherein the first group of visual media search results are responsive to a first visual media search query with text corresponding to the phrase; obtaining a respective second group of visual media search results for each of the candidate translations, wherein the second group of visual media search results are responsive to a second visual media search query with text corresponding to the candidate translation; generating a respective quality of results statistic for each of the candidate translations, wherein the quality of results statistic for each of the candidate translations is a value that represents a quality of search results that are responsive to the second visual media search query with text corresponding to the candidate translation; calculating, by one or more computers, a respective similarity score for each of the candidate translations, wherein the similarity score is an estimate of visual similarity between the first group of visual media search results and the second group of visual media search results for the candidate translation; selecting one or more of the candidate translations based on, for each candidate translation, the similarity score for the candidate translation and the quality of results statistic for the candidate translation; and associating each of the one or more selected candidate translations with the phrase as a valid translation for the phrase. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A non-transitory computer storage medium storing a program, the program comprising instructions that, when executed by a data processing apparatus, cause the data processing apparatus to perform operations comprising:
-
receiving a plurality of candidate translations for a phrase, wherein the phrase is in a first language and each candidate translation is a translation of the phrase into a different language from the first language; obtaining a first group of visual media search results, wherein the first group of visual media search results are responsive to a first visual media search query with text corresponding to the phrase; obtaining a respective second group of visual media search results for each of the candidate translations, wherein the second group of visual media search results are responsive to a second visual media search query with text corresponding to the candidate translation; generating a respective quality of results statistic for each of the candidate translations, wherein the quality of results statistic for each of the candidate translations is a value that represents a quality of search results that are responsive to the second visual media search query with text corresponding to the candidate translation; calculating a respective similarity score for each of the candidate translations, wherein the similarity score is an estimate of visual similarity between the first group of visual media search results and the second group of visual media search results for the candidate translation; selecting one or more of the candidate translations based on, for each candidate translation, the similarity score for the candidate translation and the quality of results statistic for the candidate translation; and associating each of the one or more selected candidate translations with the phrase as a valid translation for the phrase. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16)
-
-
17. A system comprising one or more computers programmed to perform operations comprising:
-
receiving a plurality of candidate translations for a phrase, wherein the phrase is in a first language and each candidate translation is a translation of the phrase into a different language from the first language; obtaining a first group of visual media search results, wherein the first group of visual media search results are responsive to a first visual media search query with text corresponding to the phrase; obtaining a respective second group of visual media search results for each of the candidate translations, wherein the second group of visual media search results are responsive to a second visual media search query with text corresponding to the candidate translation; generating a respective quality of results statistic for each of the candidate translations, wherein the quality of results statistic for each of the candidate translations is a value that represents a quality of search results that are responsive to the second visual media search query with text corresponding to the candidate translation; calculating a respective similarity score for each of the candidate translations, wherein the similarity score is an estimate of visual similarity between the first group of visual media search results and the second group of visual media search results for the candidate translation; selecting one or more of the candidate translations based on, for each candidate translation, the similarity score for the candidate translation and the quality of results statistic for the candidate translation; and associating each of the one or more selected candidate translations with the phrase as a valid translation for the phrase. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24)
-
Specification