The past few decades have been characterized by an exponential growth of digital information resources. A considerable amount of this information is semi-structured, such as XML files and metadata records and unstructured, such as scientific reports, news articles, and historical archives. These resources include a wealth of latent knowledge in a form mainly intended for human use. Semantic information elicitation refers to a set of related processes: semantic information extraction, linking, and annotation that aim to make this knowledge explicit to help computer systems make sense of the content and support ontology construction, information organization, and knowledge discovery.

In the context of GIScience research, semantic information extraction aims at processing unstructured and semi-structured resources and identifying specific types of information: places, events, topics, geospatial concepts, and relations. These may be further linked to ontologies and knowledge bases to enrich the original unstructured content with well-defined meaning, provide access to information not explicit in the original sources, and support semantic annotation and search. Semantic analysis and visualization techniques are further employed to explore aspects latent in these sources such as the historical evolution of cities, the progression of phenomena and events and people’s perception of places and landscapes.

Shape is important in GI Science because the shape of a geographical entity can have far-reaching effects on significant characteristics of that entity. In geography we are mainly concerned with two-dimensional shapes such as the outlines of islands, lakes, and administrative areas, but three-dimensional shapes may become important, for example in the treatment of landforms. Since the attribute of shape has infinitely many degrees of freedom, there can be no single numerical measure such that closely similar shapes are assigned close numerical values. Therefore different shape descriptors have been proposed for different purposes. Although it is generally desirable for a shape descriptor to be scale invariant and rotation invariant, not all proposed descriptors satisfy both these requirements. Some methods by which a shape is described using a single number are described, followed by a discussion of moment-based approaches. It is often useful to represent a complex shape by means of a surrogate shape of simpler form which facilitates storage, manipulation, and comparison between shapes; some examples of commonly used shape surrogates are presented. Another important task is to compare different shapes to determine how similar they are. The article concludes with a discussion of a number of such measures of similarity.

Social media streams have emerged as new sources to support various geospatial applications. However, traditional geospatial tools and systems lack the capacities to process such data streams, which are generated dynamically in extremely large volumes and with versatile contents. Therefore, innovative approaches and frameworks should be developed to detect an emerging event discussed over the social media, understand the extent, consequences of the event, as well as it time-evolving nature, and eventually discover useful patterns. In order to harness social media for geospatial applications, this entry introduces social media analytics technologies for harvesting, managing, mining, analyzing and visualizing the spatial, temporal, text, and network information of social media data.

This entry introduces the concept of a social network (SN), its components, and how to weight those components. It also describes some spatial properties of SNs, and how to embed SNs into GIS. SNs are graph structures that consists of nodes and edges that traditionally exist in Sociology and are newer to GIScience. Nodes typically represent individual entities such as people or institutions, and edges represent interpersonal relationships, connections or ties. Many different mathematical metrics exist to characterize nodes, edges and the larger network. When geolocated, SNs are part of a class of spatial networks, more specifically, geographic networks (i.e. road networks, hydrological networks), that require special treatment because edges are non-planar, that is, they do not follow infrastructure or form a vector on the earth’s surface. Future research in this area is likely to take advantage of 21st Century datasets sourced from social media, GPS, wireless signals, and online interactions that each evidence geolocated personal relationships.