TY - GEN
T1 - From Noisy Data to Useful Color Palettes
T2 - 18th International Conference on Information for a Better World: Normality, Virtuality, Physicality, Inclusivity, iConference 2023
AU - Cui, Hong
AU - Giebink, Noah
AU - Starr, Julian
AU - Longert, Dylan
AU - Ford, Bruce
AU - Léveillé-Bourret, Étienne
N1 - Funding Information:
Acknowledgements. The project was supported in part by NSF DBI-1661485. The authors thank anonymous Carex experts who participated in the color palette evaluation experiments.
Publisher Copyright:
© 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2023
Y1 - 2023
N2 - Due to the differences in individual’s color perception and the variations in color naming and color rendering under different settings, color has historically been a challenging trait in describing species for taxonomic and systematic research. Reusing a noisy color dataset collected from high-quality images of Carex specimens, we developed a data mining method (e.g., clustering and classification) for constructing domain-specific color palettes. Color palettes associated with color values measured in a color space help systematists record color data in a way that the differences in colors can be more accurately compared and computed, making color data interoperable and reusable. The Carex color palette was evaluated by Carex experts and the evaluation data showed that experts overwhelmingly preferred using color palette over color strings.
AB - Due to the differences in individual’s color perception and the variations in color naming and color rendering under different settings, color has historically been a challenging trait in describing species for taxonomic and systematic research. Reusing a noisy color dataset collected from high-quality images of Carex specimens, we developed a data mining method (e.g., clustering and classification) for constructing domain-specific color palettes. Color palettes associated with color values measured in a color space help systematists record color data in a way that the differences in colors can be more accurately compared and computed, making color data interoperable and reusable. The Carex color palette was evaluated by Carex experts and the evaluation data showed that experts overwhelmingly preferred using color palette over color strings.
KW - Automate color palette creation
KW - Color traits
KW - Data mining
KW - Inconsistent color data
KW - K-means
KW - L a b color space
KW - Support vector machines
KW - sRGB color space
KW - t-SNE
UR - http://www.scopus.com/inward/record.url?scp=85151056894&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85151056894&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-28035-1_35
DO - 10.1007/978-3-031-28035-1_35
M3 - Conference contribution
AN - SCOPUS:85151056894
SN - 9783031280344
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 469
EP - 481
BT - Information for a Better World
A2 - Sserwanga, Isaac
A2 - Goulding, Anne
A2 - Moulaison-Sandy, Heather
A2 - Du, Jia Tina
A2 - Soares, António Lucas
A2 - Hessami, Viviane
A2 - Frank, Rebecca D.
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 13 March 2023 through 17 March 2023
ER -