The art market is a large and growing part of the global economy. However, uncertainty about prices, which can be problematic for many shareholders, can inhibit the growth of this market. This work discusses methods for the development of a knowledge based recommender system that will price contemporary fine art. Artworks are unique and often rare purchases. This makes a knowledge based system particularly suitable for that problem area. To the knowledge of this researcher, there are no knowledge based recommender systems for artwork pricing currently available. In this dissertation, I will discuss past research in the field of art analytics, and the competing factors which drive art prices. I will also discuss the dataset that has been collected for use on this project. I will then discuss the development of both visual and textual features for this recommender system. Methods for the clustering of artists using visual features will be discussed. This work will also include an exploration of the development of personalized models based on these artist clusters and discuss their impact on the efficacy of the models built. Lastly, this work will discuss a final structure for a recommender system and how it could be created moving forward.