logo
ResearchBunny Logo
Explainable dimensionality reduction (XDR) to unbox AI ‘black box’ models: A study of AI perspectives on the ethnic styles of village dwellings

Interdisciplinary Studies

Explainable dimensionality reduction (XDR) to unbox AI ‘black box’ models: A study of AI perspectives on the ethnic styles of village dwellings

X. Li, D. Chen, et al.

Discover a revolutionary explainable Dimensionality Reduction (XDR) framework that transforms high-dimensional AI knowledge into clear insights! With a compelling case study on ethnic styles of village dwellings in Guangdong, China, led by Xun Li, Dongsheng Chen, Weipan Xu, Haohui Chen, Junjun Li, and Fan Mo, this research highlights pivotal features that enhance our understanding of culture and architecture.

00:00
00:00
Playback language: English
Abstract
This paper proposes a novel explainable Dimensionality Reduction (XDR) framework to translate high-dimensional tacit knowledge learned by AI into human-understandable explicit knowledge. A case study on recognizing ethnic styles of village dwellings in Guangdong, China, using a Mask R-CNN model, reveals patio, size, length, direction, and asymmetric shape as key distinguishing features for Canton, Hakka, and Teochew styles. Results align with existing field studies and uncover evidence of Hakka migration, demonstrating XDR's potential for expanding domain knowledge.
Publisher
Humanities and Social Sciences Communications
Published On
Jan 26, 2023
Authors
Xun Li, Dongsheng Chen, Weipan Xu, Haohui Chen, Junjun Li, Fan Mo
Tags
Dimensionality Reduction
Explainable AI
Ethnic styles
Village dwellings
Canton
Hakka
Teochew
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny