OccGen: Selection of Real-world Multilingual Parallel Data Balanced in Gender within Occupations

Marta R. Costa-jussà, Christine Basta, Oriol Domingo, André Niyongabo Rubungo

Keywords: Balanced Multilingual Data Set, Gender, Occupations, Machine Translation

TL;DR: We present the OccGen toolkit that builds multilingual parallel data sets balanced in gender within occupations. The toolkit is released together with two datasets in four high-resource languages and in a low-resource language (with English).

Abstract:

This paper describes the OCCGEN toolkit, which allows extracting multilingual parallel data balanced in gender within occupations. OCCGEN can extract datasets that reflect gender diversity (beyond binary) more fairly in society to be further used to explicitly mitigate occupational gender stereotypes. We propose two use cases that extract evaluation datasets for machine translation in four high-resource
languages from different linguistic families and in a low-resource African language. Our analysis of these use cases shows that translation outputs in high-resource languages tend to worsen in feminine subsets (compared to masculine). This can be explained because less attention is paid to the source sentence. Then, more attention is given to the target prefix overgeneralizing to the most frequent masculine forms.

Author Statement: Yes

URL: https://github.com/mt-upc/OccGen_dataset

Dataset Url: https://github.com/mt-upc/OccGen_dataset

License: CC-BY-SA 3.0

Contribution Process Agreement: Yes

In Person Attendance: Yes