Methods Of Avoiding Bias In Your Translations Machine-learning-based discrimination is an issue that has been ongoing, even when directly addressed by the largest and most sophisticated companies in the world. Skip ahead to September, and published a piece to show that Google Translate is still systematically changing the gender of translations when they "do not fit with stereotypes." Going into 2021, it is hard to imagine this issue has been fully solved or will be in the near future. However, in April 2020, the problem persisted, and Google once again was forced to directly address the issue. In 20, Google Translate again faced heavy criticism for perceived sexism, something the company promised to address. Google Translate, the most prolific of the machine-based translation tools, has struggled with various forms of discrimination in its translations as far back as 2013. This discrimination issue extends to translations as well. Even the data used for machine learning itself can be biased, depending on how it was collected and the weight with which each piece of data is assigned.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |