 This paper proposes a knowledge-based approach to represent geo-visualization knowledge using semantic web technologies, which enables inference for deriving corresponding geometries and portrayals for visualization under different conditions. A prototype system is developed that uses geospatial link data as underlying data and formalizes some geo-visualization knowledge into a knowledge base to visualize the data and provide rich semantics to users, partially forming the foundation for the vision of web of knowledge for geo-visualization. This article was authored by Wei Minghuang and Lars Harry.