What is it about?

This study proposes a search-engine-based color palette generator inspired by a Natural Language Processing algorithm that filters and clusters semantically related words and colors. TF-IDF and Word2Vec algorithms for processing textual data have been adapted to handle color data extracted from image search results.

Featured Image

Why is it important?

Our new design tool computes color palettes for any textual input as a design concept, whose association is not only stronger than traditional crowd-sourced dictionaries but can be tuned to various demographic factors such as artistic training and cultural backgrounds.


Our data-based approach suggests a fundamentally new way to the existing method of color planning, which largely depends on human intuition.

Sangwon Lee
Yonsei University

Read the Original

This page is a summary of: Color2Vec: Web-based Modeling of Word-color Association with Sociocultural Contexts, ACM Transactions on Computer-Human Interaction, November 2022, ACM (Association for Computing Machinery), DOI: 10.1145/3571816.
You can read the full text:




The following have contributed to this page