Share Email Print
cover

Proceedings Paper • new

deb2viz: Debiasing gender in word embedding data using subspace visualization
Author(s): Enoch Opanin Gyamfi; Yunbo Rao; Miao Gou; Yanhua Shao
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Word embedding have been used in numerous Natural Language Processing and Machine Learning tasks. However, it is a high-dimensional vector field that propagate stereotypes to software applications. Its current debiasing frameworks do not completely capture its embedded patterns. In this paper, we propose deb2viz, a visual debiasing approach that explores and manipulates the high-dimensional patterns of word embedding field. First, we partition this vector field into interrelated low-dimensional subspaces to equalize and neutralize distances between gender-definitional and gender-neutral words. To further reduce gender bias, we update the distances of appropriate nearest neighbors for gender-neutral words to be arbitrarily close. Experimental results on several benchmark standards show the competitiveness of our proposed method in mitigating bias within pre-trained word2vec embedding model.

Paper Details

Date Published: 3 January 2020
PDF: 8 pages
Proc. SPIE 11373, Eleventh International Conference on Graphics and Image Processing (ICGIP 2019), 113732F (3 January 2020); doi: 10.1117/12.2557465
Show Author Affiliations
Enoch Opanin Gyamfi, Univ. of Electronic Science and Technology of China (China)
Yunbo Rao, Univ. of Electronic Science and Technology of China (China)
Miao Gou, Univ. of Electronic Science and Technology of China (China)
Yanhua Shao, Southwest Univ. of Science and Technology (China)


Published in SPIE Proceedings Vol. 11373:
Eleventh International Conference on Graphics and Image Processing (ICGIP 2019)
Zhigeng Pan; Xun Wang, Editor(s)

© SPIE. Terms of Use
Back to Top