Ranking Buildings and Mining the Web for Popular Architectural Patterns

Abstract

Knowledge about the reception of architectural structures is crucial for architects and urban planners. Yet obtaining such information has been a challenging and costly activity. However, with the advent of the Web, a vast amount of structured and unstructured data describing architectural structures has become available publicly. This includes information about the perception and use of buildings (for instance, through social media), and structured information about the building's features and characteristics (for instance, through public Linked Data). Hence, first mining (i) the popularity of buildings from the social Web and (ii) then correlating such rankings with certain features of buildings, can provide an efficient method to identify successful architectural patterns. In this paper we propose an approach to rank buildings through the automated mining of Flickr metadata. By further correlating such rankings with building properties described in Linked Data we are able to identify popular patterns for particular building types (airports, bridges, churches, halls, and skyscrapers). Our approach combines crowdsourcing with Web mining techniques to establish influential factors, as well as ground truth to evaluate our rankings. Our extensive experimental results depict that methods tailored to specific structure types allow an accurate measurement of their public perception.