Exploring Similarity - Improving Product Search with Parallel Coordinates.

Faceted browsing is an established and well-known paradigm for product search. However, if the user is unfamiliar with the topic and the provided facets, he may not be able to sufficiently reduce the amount of results. In order to increase the understanding of the bidirectional relation between facets and result set, we present an interface concept that allows manifold approaches for product search, analysis and comparison starting with a single product or a summarizing visualization of the entire data set. Moreover, various product features can be analyzed in order to support decision-making. Even without detailed knowledge of a specific topic, the user is able to estimate the range and distribution of characteristics in relation to known or desired features. Conventional list-based search forms do not provide such a quick overview. Our concept is based on two visualization techniques that allow the representation of multi-dimensional data across a set of parallel axes: parallel coordinates and parallel sets.