Systems and methods for determining translation accuracy in multi-user multi-lingual communications
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 accessing the accuracy of translations using word based and language based features are described.
1. A method comprising:
performing by one or more computer processors:
obtaining an original text message in a first language authored by a first user;
obtaining a translation of the original text message in a second language;
comparing a plurality of features associated with the original text message with a respective plurality of features associated with the translation, the plurality of features comprising (i) at least one word-based feature selected from the group consisting of a word count, a character count, an emoji, a number, and a punctuation mark, and (ii) at least one language-based feature comprising a part of speech selected from group consisting of verbs, nouns, adverbs, and adjectives;
calculating a metric based on the comparison of the features, and
determining an accuracy of the translation based on the calculated metric.
2. The method of claim 1 , wherein the translation is authored by a second user.
3. The method of claim 1 , wherein the translation comprises a correction of a prior translation of the original text message.
4. The method of claim 1 , wherein comparing the plurality of features comprises determining a difference in the at least one word-based feature between the original text message and the translation.
5. The method of claim 1 , wherein comparing the plurality of features comprises determining a difference in a number of occurrences of the part of speech between the original text message and the translation.
6. The method of claim 1 , wherein calculating the metric comprises determining an alignment of at least one word in the original text message with at least one respective word in the translation.
7. The method of claim 1 , wherein the metric is further based on grammar of the translation.
8. The method of claim 1 , wherein calculating the metric comprises:
generating a part-of-speech n-gram representation for the translation; and
computing a probability of the n-gram representation for the second language.
9. The method of claim 1 , wherein calculating the metric comprises fitting the features with a linear regression model.
10. The method of claim 1 , wherein calculating the metric comprises fitting the features with one or more machine learning algorithms.
11. 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:
obtaining an original text message in a first language authored by a first user;
obtaining a translation of the original text message in a second language;
comparing a plurality of features associated with the original text message with a respective plurality of features associated with the translation, the plurality of features comprising (i) at least one word-based feature selected from the group consisting of a word count, a character count, an emoji, a number, and a punctuation mark, and (ii) at least one language-based feature comprising a part of speech selected from group consisting of verbs, nouns, adverbs, and adjectives;
calculating a metric based on the comparison of the features, and
determining an accuracy of the translation based on the calculated metric.
12. The system of claim 11 , wherein the translation is authored by a second user.
13. The system of claim 11 , wherein the translation comprises a correction of a prior translation of the original text message.
14. The system of claim 11 , wherein comparing the plurality of features comprises determining a difference in the at least one word-based feature between the original text message and the translation.
15. The system of claim 11 , wherein comparing the plurality of features comprises determining a difference in a number of occurrences of the part of speech between the original text message and the translation.
16. The system of claim 11 , wherein calculating the metric comprises determining an alignment of at least one word in the original text message with at least one respective word in the translation.
17. The system of claim 11 , wherein the metric is further based on grammar of the translation.
18. The system of claim 11 , wherein calculating the metric comprises:
generating a part-of-speech n-gram representation for the translation; and
computing a probability of the n-gram representation for the second language.
19. The system of claim 11 , wherein calculating the metric comprises fitting the features with a linear regression model.
20. The system of claim 11 , wherein calculating the metric comprises fitting the features with one or more machine learning algorithms.
21. A manufacture comprising:
non-transitory computer readable media comprising executable instructions, the executable instructions being executable by one or more processors to perform operations comprising:
obtaining an original text message in a first language authored by a first user;
obtaining a translation of the original text message in a second language;
comparing a plurality of features associated with the original text message with a respective plurality of features associated with the translation, the plurality of features comprising (i) at least one word-based feature selected from the group consisting of a word count, a character count, an emoji, a number, and a punctuation mark, and (ii) at least one language-based feature comprising a part of speech selected from group consisting of verbs, nouns, adverbs, and adjectives;
calculating a metric based on the comparison of the features, and
determining an accuracy of the translation based on the calculated metric.