Systems and methods for multi-user multi-lingual communications

• US 9,881,007 B2
• Filed: 04/06/2015
• Issued: 01/30/2018
• Est. Priority Date: 02/08/2013
• Status: Expired due to Fees

• Current Assignee: Mz Ip Holdings, LLC (AppLovin Corporation)
• Original Assignee: Machine Zone, Inc. (AppLovin Corporation)
• Inventors: Orsini, Francois, Bojja, Nikhil
• Primary Examiner(s): SPOONER, LAMONT M
• Application Number:
  US14/679,224
• Publication Number:
  US 20150213008A1
• Time in Patent Office: 1,030 Days
• Field of Search: 704 2- 8
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06F 40/51 Translation evaluation
• 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
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: G06Q 30/0217 involving input on products or services in exchange for incentives or rewards
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems
• CPC Class Codes: G10L 13/08 Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence

Abstract

Various embodiments described herein facilitate multi-lingual communications. The systems and methods of some embodiments enable multi-lingual communications through different modes of communication including, for example, Internet-based chat, e-mail, text-based mobile phone communications, postings to online forums, postings to online social media services, and the like. Certain embodiments implement communication systems and methods that translate text between two or more languages. Users of the systems and methods may be incentivized to submit corrections for inaccurate or erroneous translations, and may receive a reward for these submissions. Systems and methods for assessing the accuracy of translations are described.

Claims

1. A computer-implemented method comprising:

sending each proposal of a plurality of proposals at a different time to a client device of a user, the proposal comprising a respective request for the user to translate a respective text message and a respective incentive for the translation;

receiving from the client device a plurality of translations authored by the user, each translation being associated with one of the proposals and being a translation of the text message of the associated proposal;

for each proposal,

comparing a plurality of features associated with the text message with a respective plurality of features associated with the translation to generate respective feature scores, the features comprising word-based features, language-based features, word alignment features, and combinations thereof, and

calculating an accuracy based on a weighted combination of the respective feature scores wherein the weights are derived using a statistical regression to model accuracy of translations;

identifying a deviation from a norm in user translation accuracy over a period of time based on the calculated accuracies;

updating a confidence score for the user based on the deviation in user translation accuracy; and

revoking the translation privileges of the user when the confidence score does not satisfy a threshold value.

2. The method of claim 1, wherein the user is a participant in an online game, and wherein the respective incentive comprises at least one of a virtual good and a virtual currency for the online game.

3. The method of claim 1, wherein identifying a deviation in user translation accuracy comprises using item response theory.

4. The method of claim 1, further comprising:

for a first translation of the translations, rewarding the user with the incentive of the proposal associated with the first translation when the first translation is determined to be accurate.

5. The method of claim 1, wherein the word-based features comprise at least one of a word count, a character count, an emoji, a number, and a punctuation mark.

6. The method of claim 1, wherein the language-based features comprise at least one of verbs, nouns, adverbs, and adjectives.

7. The method of claim 6, wherein comparing the plurality of features comprises comparing a number of words having the part of speech in the text message with a number of words having the part of speech in the translation.

8. The method of claim 1, wherein the word-alignment features comprise an alignment of at least one word in the text message with at least one respective word in the translation.

9. A system comprising:

a non-transitory computer readable medium having instructions stored thereon; and

at least one processor configured to execute the instructions to perform operations comprising;

sending each proposal of a plurality of proposals at a different time to a client device of a user, the proposal comprising a respective request for the user to translate a respective text message and a respective incentive for the translation;

receiving from the client device a plurality of translations authored by the user, each translation being associated with one of the proposals and being a translation of the text message of the associated proposal;

for each proposal,

comparing a plurality of features associated with the text message with a respective plurality of features associated with the translation to generate respective feature scores, the features comprising word-based features, language-based features, word alignment features, and combinations thereof, and

calculating an accuracy based on a weighted combination of the respective feature scores wherein the weights are derived using a statistical regression to model accuracy of translations;

identifying a deviation from a norm in user translation accuracy over a period of time based on the calculated accuracies;

updating a confidence score for the user based on the deviation in user translation accuracy; and

revoking the translation privileges of the user when the confidence score does not satisfy a threshold value.

10. The system of claim 9, wherein the user is a participant in an online game, and wherein the respective incentive comprises at least one of a virtual good and a virtual currency for the online game.

11. The system of claim 9, wherein identifying a deviation in user translation accuracy comprises using item response theory.

12. The system of claim 9, further comprising:

for a first translation of the translations, rewarding the user with the incentive of the proposal associated with the first translation when the first translation is determined to be accurate.

13. The system of claim 9, wherein the word-based features comprise at least one of a word count, a character count, an emoji, a number, and a punctuation mark.

14. The system of claim 9, wherein the language-based features comprise at least one of verbs, nouns, adverbs, and adjectives.

15. The system of claim 14, wherein comparing the plurality of features comprises comparing a number of words having the part of speech in the text message with a number of words having the part of speech in the translation.

16. The system of claim 9, wherein the word-alignment features comprise an alignment of at least one word in the text message with at least one respective word in the translation.

17. An article comprising:

non-transitory computer readable media comprising executable instructions, the executable instructions being executable by one or more processors to perform operations comprising;

sending each proposal of a plurality of proposals at a different time to a client device of a user, the proposal comprising a respective request for the user to translate a respective text message and a respective incentive for the translation;

receiving from the client device a plurality of translations authored by the user, each translation being associated with one of the proposals and being a translation of the text message of the associated proposal;

for each proposal,

comparing a plurality of features associated with the text message with a respective plurality of features associated with the translation to generate feature scores, the features comprising word-based features, language-based features, word alignment features, and combinations thereof, and

calculating an accuracy based on a weighted combination of the respective feature scores wherein the weights are derived using a statistical regression to model accuracy of translations;

identifying a deviation from a norm in user translation accuracy over a period of time based on the calculated accuracies;

updating a confidence score for the user based on the deviation in user translation accuracy; and

revoking the translation privileges of the user when the confidence score does not satisfy a threshold value.