id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_b7cwv4jn75bozbnft4vlwqq5cy	William N. Goetzmann	Beauty is in the bid of the beholder: An empirical basis for style	2016	32	.pdf	application/pdf	9539	1094	77	We develop a method for classification of works of art based on their price dynamics. developed clustering algorithm to group artists that represent similar styles. empirically on a ten-year sample of price data for paintings by 58 artists. the aim is to cluster artists with similar styles by observing price dynamics of sold works of art. Hierarchical clustering algorithms can work either bottom-up, by agglomerating points and Figure 3: Tree built by hierarchical clustering algorithm with cosine distance and complete similarity Figure 4: Clusters identified with Laplacian eigenmap algorithm. would also like an "objective" measure of how "similar" artists within a cluster are, and how Number of artists in each cluster that represent different styles. be any relationship between style and the clusters identified by the hierarchical tree algorithm. and we could reject that these clusters were random with respect to style and age of artist.	./cache/work_b7cwv4jn75bozbnft4vlwqq5cy.pdf	./txt/work_b7cwv4jn75bozbnft4vlwqq5cy.txt