METHOD AND APPARATUS FOR ASSESSING A TRANSLATION
Methods, apparatus and computer program products are provided in order to assess a translation following performance of the translation. The methods, apparatus and computer program products may determine input segments of a source language document that may prove to be problematic from a translatability standpoint, such as the input segments of the source language document that may have multiple output variants. As such, methods, apparatus and computer program products may provide feedback to the author or owner of the source language document that may influence the generation of subsequent source language documents so as to have improved translatability.
1 . A method of assessing a translation comprising:
aligning, with a processor, input segments of a source language document with corresponding output segments of a target language document;
for each input segment, identifying variations between the output segments corresponding to a respective input segment, wherein identifying the variations comprises identifying a reference translation and one or more output variants for the respective input segment; and
determining the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
2 . A method according to claim 1 further comprising providing feedback regarding the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
3 . A method according to claim 1 wherein identifying the reference translation comprises identifying the output segment that most frequently corresponds to the respective input segment.
4 . A method according to claim 1 wherein determining the one or more input segments having corresponding output variants that fail to satisfy the control limit for translation variation comprises determining a measurement of similarity between each output variant and the reference translation.
5 . A method according to claim 4 wherein determining the measurement of similarity comprises determining a longest common subsequence between each output variant and the reference translation.
6 . A method according to claim 5 wherein determining the measurement of similarity comprises determining a similarity metric based upon recall and precision of the longest common subsequence between each output variant and the reference translation.
7 . A method according to claim 6 wherein the control limit is based upon the similarity metric.
8 . A computing device configured to assess a translation, wherein the computing device comprises a processor configured to align input segments of a source language document with corresponding output segments of a target language document, wherein the processor is also configured, for each input segment, to identify variations between the output segments corresponding to a respective input segment including identification of a reference translation and one or more output variants for the respective input segment, and wherein the processor is configured to determine the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
9 . A computing device according to claim 8 wherein the processor is further configured to provide feedback regarding the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
10 . A computing device according to claim 8 wherein the processor is configured to identify the reference translation by identifying the output segment that most frequently corresponds to the respective input segment.
11 . A computing device according to claim 8 wherein the processor is configured to determine the one or more input segments having corresponding output variants that fail to satisfy the control limit for translation variation by determining a measurement of similarity between each output variant and the reference translation.
12 . A computing device according to claim 11 wherein the processor is configured to determine the measurement of similarity by determining a longest common subsequence between each output variant and the reference translation.
13 . A computing device according to claim 12 wherein the processor is configured to determine the measurement of similarity by determining a similarity metric based upon recall and precision of the longest common subsequence between each output variant and the reference translation.
14 . A computing device according to claim 13 wherein the control limit is based upon the similarity metric.
15 . A computer program product for assessing a translation and comprising at least one computer-readable storage medium having computer-executable program code portions stored therein, the computer-executable program code portions comprising:
program code instructions for aligning input segments of a source language document with corresponding output segments of a target language document;
for each input segment, program code instructions for identifying variations between the output segments corresponding to a respective input segment, wherein identifying the variations comprises identifying a reference translation and one or more output variants for the respective input segment; and
program code instructions for determining the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
16 . A computer program product according to claim 15 further comprising program code instructions for roviding feedback regarding the one or more input segments having corresponding output variants that fail to satisfy a control limit for translation variation.
17 . A computer program product according to claim 15 wherein the program code instructions for identifying the reference translation comprise program code instructions for identifying the output segment that most frequently corresponds to the respective input segment.
18 . A computer program product according to claim 15 wherein the program code instructions for determining the one or more input segments having corresponding output variants that fail to satisfy the control limit for translation variation comprise program code instructions for determining a measurement of similarity between each output variant and the reference translation.
19 . A computer program product according to claim 18 wherein the program code instructions for determining the measurement of similarity comprise program code instructions for determining a longest common subsequence between each output variant and the reference translation.
20 . A computer program product according to claim 5 wherein the program code instructions for determining the measurement of similarity comprise program code instructions for determining a similarity metric based upon recall and precision of the longest common subsequence between each output variant and the reference translation, wherein the control limit is based upon the similarity metric.