Automatic sketch map creation from labeled planar graph

Abstract

Maps constructed in Euclidean space are commonly used to visuallypresent information about the real world. However, their creation isresource intensive, be it financial, technical, human, or time-con-suming, which can limit their timeliness and detail. A simpler formof visualization of data about the real world is represented bysketch maps, which mainly capture the topology and mutual fea-tures’ spatial location. By default, they are drawn by hand. This pre-supposes that the creator has a good knowledge of the depictedterritory, can create a cognitive map, and is skilled in transformingit into a graphical form. Sketch maps can be detailed and up-to-date if these prerequisites are met. Our question was whether it ispossible to meet these assumptions in another way: acquire know-ledge of the territory by processing narratives related to the area ofinterest, create a suitable computer representation for further proc-essing, and automatically generate the resulting sketch map. Thisarticle presents the last step – creating a sketch map based on spa-tial data acquired from narratives. The results show that even with-out metric data, it is possible to automatically generate a sketchmap visually close to the actual situation.