Map Based Visualization of Product Catalogs

Traditionally, recommender systems present recommendations in ranked lists to the user. In contentand knowledge-based recommender systems, these lists are often sorted on some notion of similarity with a query, ideal product specification, or sample product. However, a lot of information is lost in this way, since two products with the same similarity to a query can differ from this query on a completely different set of product characteristics. When using a two dimensional map based visualization of the recommendations, it is possible to retain part of this information. In the map, we can then position recommendations that are similar to each other in the same area of the map. A domain in which this map based approach can be very useful is electronic commerce, since electronic commerce stores sell a large number of products, often described by a large number of characteristics, but that are, in certain product categories, easily recognizable by an image of the product at hand. E-commerce domains for which this holds are, for example, consumer electronics and real estate. In the industry, one tries nowadays to improve usability of this kind of websites, using a trend called visual shopping. An example is CrispyShop.com1, which compares products on one characteristic and price using a chart. Another approach taken in visual shopping is selecting items based on visual similarity (like color and shape), such as is done by Like.com2 and Modista3 which is especially useful in fashion related fields. An approach more closely related to the map based approach, is taken