Kwak et al., "FrameAxis: characterizing microframe bias and intensity with word embedding" (2021)
2021-07-22 → 2026-08-14
Haewoon Kwak, Jisun An, Elise Jing, and Yong-Yeol Ahn, PeerJ Computer Science 7, e644 (2021)
Link | arXiv | PDF
@article{kwak2021frameaxis,
author = {Haewoon Kwak and Jisun An and Elise Jing and Yong-Yeol Ahn},
title = {FrameAxis: characterizing microframe bias and intensity with word embedding},
journal = {PeerJ Computer Science},
volume = {7},
pages = {e644},
doi = {10.7717/peerj-cs.644},
archivePrefix = {arXiv},
eprint = {2002.08608},
primaryClass = {cs.CL},
year = {2021},
}
Introduces FrameAxis, an unsupervised method that uses semantic axes in word embeddings to measure which microframes a document emphasizes and in which direction. It separates microframe bias from intensity and provides word- and document-level explanations.
Framing, Semantic axis, Word embedding