Providing human-generated and machine-generated trusted translations
Customers having a translation project to select a translation method from a variety of options, ranging from a completely human translation to a completely automated translation. For human translations, translation job information may be communicated through one or more network service modules which execute within a network service application, such as a web-based networking application. A network service module may register a user having an account with the network service application as a translator and communicate translation jobs to the user. One or more users who express interest in performing the translation are selected to perform a translation job, each job comprising at least a portion of the translation project. After a user provides a translation for the translation job, the translation is analyzed to generate a trust level prediction for the translation. A user translation profile may be updated after each translation to reflect the user's performance.
1. A method for providing a trusted translation, comprising:
receiving a human-generated translation of a document from a source language to a target language, the human-generated translation being produced by a person;
generating a trust level prediction of the human-generated translation based at least in part on a mapping by executing a quality-prediction engine stored in memory, the trust level prediction of the human-generated translation associated with translational accuracy of the human-generated translation, the quality-prediction engine being calibrated, the calibration including:
obtaining a plurality of opinions for a plurality of sample translations performed by the person, each of the opinions from a human and indicating a perceived trust level of corresponding sample translations, and
using the quality prediction-engine to determine a trust level of each of the plurality of sample translations,
determining a relationship between the plurality of opinions and the trust levels of each of the plurality of sample translations, and
tuning the mapping to minimize any difference between the plurality of opinions and the trust levels of each of the plurality of sample translations; and
outputting the human-generated translation and the trust level prediction of the human-generated translation.
2. The method of claim 1 , further comprising providing an interface configured to receive a translation selection of one of a plurality of translation methods, the plurality of translation methods including both a human translation method and a machine translation method.
3. The method of claim 1 , wherein outputting includes reporting information based on the trust level prediction to a user that generated the translation.
4. The method of claim 1 , wherein outputting includes reporting information based on the trust level prediction to a module executing within a remote application.
5. The method of claim 1 , further comprising transmitting translation information to a module executing within a remote application.
6. The method of claim 5 , wherein the translation information includes a description of the translation to be performed.
7. The method of claim 5 , wherein the translation information includes information for initiating access to the content to be translated.
8. The method of claim 1 , further comprising receiving configuration information for a translation to be performed by humans.
9. The method of claim 8 , wherein the configuration information includes a selection of one or more modules from a plurality of modules within one or more remote applications.
10. The method of claim 9 , wherein the remote applications are network service applications executing on remote application servers.
11. A non-transitory computer readable storage medium having embodied thereon a program, the program being executable by a processor to perform a method for providing a trusted translation, the method comprising:
receiving a human-generated translation of a document from a source language to a target language, the human-generated translation being produced by a person;
generating a trust level prediction of the human-generated translation based at least in part on a mapping, the trust level associated with translational accuracy of the human-generated translation;
calibrating, the calibrating including:
obtaining a plurality of opinions for a plurality of sample translations performed by the person, each of the opinions from a human and indicating a perceived trust level of corresponding sample translations, and
using the quality prediction-engine to determine a trust level of each of the plurality of sample translations,
determining a relationship between the plurality of opinions and the trust levels of each of the plurality of sample translations, and
tuning the mapping to minimize any difference between the plurality of opinions and the trust levels of each of the plurality of sample translations; and
outputting the human-generated translation and the trust level prediction of the human-generated translation.
12. The non-transitory computer readable storage medium of claim 11 , the method further comprising providing an interface configured to receive a translation selection of one of a plurality of translation methods, the plurality of translation methods including both a human translation method and a machine translation method.
13. The non-transitory computer readable storage medium of claim 11 , wherein outputting includes reporting information based on the trust level prediction to a user that generated the translation.
14. The non-transitory computer readable storage medium of claim 11 , wherein outputting includes reporting information based on the trust level prediction to a module executing within a remote application.
15. The non-transitory computer readable storage medium of claim 11 , the method further comprising transmitting translation information to a module executing within a remote application.
16. The non-transitory computer readable storage medium of claim 15 , wherein the translation information includes a description of the translation to be performed.
17. The non-transitory computer readable storage medium of claim 15 , wherein the translation information includes information for initiating access to the content to be translated.
18. The non-transitory computer readable storage medium of claim 11 , the method further comprising receiving configuration information for a translation to be performed by humans.
19. The non-transitory computer readable storage medium of claim 18 , wherein the configuration information includes a selection of one or more modules from a plurality of modules within one or more remote applications.
20. The non-transitory computer readable storage medium of claim 19 , wherein the remote applications are network service applications executing on remote application servers.
21. The method of claim 1 , further comprising:
receiving user information from a user having an account with a network service by executing a network service module stored in memory;
receiving translation job information from a remote application by the network service module; and
providing the translation job information to the user.
22. The method of claim 21 , wherein the user information includes user translation information.
23. The method of claim 21 , wherein receiving user information includes registering the user with the network service module.
24. The method of claim 21 , further comprising communicating the translation job information to users of the network service for which user information was received.
25. The method of claim 21 , further comprising receiving a request from one or more users to perform the translation by the network service module.
26. The method of claim 21 , further comprising selecting one or more users to perform the translation by the network service module.
27. The method of claim 21 , further comprising receiving trust score information regarding a translation performed by a user.