Art I Don't Like

A web-based interactive art experience that provides personalized content to users

Team Lead: Sarah Frost
Additional Team Members: Manu Mathew Thomas
Art I Don’t Like is a web-based interactive art experience that provides personalized content to users. The user is presented with a set of art images and is prompted to choose the pieces that they find visually appealing. Our “anti-recommender” algorithm analyses the selected images and computes similarity scores. The system then shows the user a work of art with a low similarity score, that is, a work that the user might not like. The underlying technology is a neural network that classifies the art based on artist and genre. Future work will include other characteristics. This system will expose users to a broad range of art and shed light on how recommendations are made. Users will be encouraged to leave feedback about why they like or dislike the art they have been shown. Art I Don’t Like gives users a digital space to view and interact with art that they may not be familiar with. This is an example of a recommender system that has been expanded to allow for serendipity and exploration.

Our grant from the 2018-2019 IDEA Hub grant will allow us to present this project at the Museums and the Web conference, as well as hire a web designer to provide feedback on the visual design of our project. In addition, this grant will fund the web infrastructure needed to keep the site running online. Finally, we will be able to pay for Search-Engine Optimization, which will increase traffic to our site and allow more people to use our tool.

The project is being developed at www.artidontlike.com