Genes - Master's thesis

[Lukáš Marek]

Dear Artsy API Development Team,

my name is Lukáš Marek, and I am a fine art master's student currently working on my thesis, titled "Approaches to Artistic Creation". I'm writing to you today with some questions regarding the data available through the Artsy API, as it relates to my research. Your platform and the Art Genome Project are incredibly valuable resources, and I'm hoping you can clarify a few points for me.

I've been reading about the Art Genome Project, specifically the article The Art Genome Project: Seven Facts About the Art Genome Project, which describes the difference between genes (with a scalar value 0-100) and tags (which are binary). However, in my initial exploration of the API, I haven't been able to find these scalar values for genes associated with artists, nor have I found any data related to tags.

My specific questions are:

1. Gene Scalars: Is it possible to access the 0-100 scalar value associated with each artist's gene via the API?
2. Tags: The article mentions over 12,000 binary tags. Is there any way to access these tags through the API?
3. Categories on Artist Pages: On artist pages, such as Andy Warhol's about page, there is a list of "categories" (24 in Warhol's case). The API seems to return a much smaller number of genes (5 for Warhol). Are these "categories" a combination of genes and tags, or are they something entirely different? Is there a way to access this complete list of categories for each artist, either through the API or another method? If not via API, would web scraping this information for personal, educational use be permissible?

The goal of my thesis is to anchor my own approach to artistic creation among approaches of established artists. This will help define my approach, possibly even name it, and prepare the groundwork for further research and development of it in my doctoral studies. My plan is to create locally on my computer a representation of my artistic profile by manually assigning myself genes, tags, and/or categories (depending on what data is available). I would then use algorithms like "k-nearest neighbors" to identify similar/closest artists to myself, based on the vector distances. The scalar values associated with genes would improve the precision of my analysis. All this is purely for my master's thesis and potentially for mentioned future doctoral studies.

I understand that API access and data availability may be subject to limitations, but any information or guidance you could provide regarding these questions would be immensely helpful.

Sincerely,
Lukáš Marek