Abstract
In a typical data collection process, a surveyed spatial object is annotated upon creation, and is classified based on its attributes. This annotation can also be guided by textual definitions of objects. However, interpretations of such definitions may differ among people, and thus result in subjective and inconsistent classification of objects. This problem becomes even more pronounced if the cultural and linguistic differences are considered. As a solution, this paper investigates the role of topology as the defining characteristic of a class of spatial objects. We propose a data mining approach based on frequent itemset mining to learn patterns in topological relations between objects of a given class and other spatial objects. In order to capture topological relations between more than two (linear) objects, this paper further proposes an extension of the 9-intersection model for topological relations of line geometries. The discovered topological relations form topological constraints of an object class that can be used for spatial object classification. A case study has been carried out on bridges in the OSM dataset for the state of Victoria, Australia. The results show that the proposed approach can successfully learn topological constraints for the class bridge, and that the proposed extended topological model for line geometries outperforms the 9-intersection model in this task.
Highlights
Responsive maintenance of spatial data and its quality is becoming difficult due to the increase of the volume of collected data and its heterogeneity
This paper has addressed the problem of object classification in spatial datasets
It presents a departure from the majority of previous work where objects were classified based on attribute annotations of spatial features. Since such classification is often guided by textual definitions of the attributes and their values, and performed by humans, it is prone to subjectivity and vague definitions, leading to classification inaccuracies
Summary
Responsive maintenance of spatial data and its quality is becoming difficult due to the increase of the volume of collected data and its heterogeneity. The association of a surveyed or manually vectorized spatial object with an object class is performed by a human operator, interpreting the definition and deciding which class the object belongs to. This may happen during data collection (e.g., a surveyor annotating the measured geometries) or in a separate, consecutive process. Consider the definition of a bridge from the Merriam-Webster dictionary: “a structure carrying a pathway or roadway over a depression or obstacle (such as a river).” Such a classification is problematic because it depends on human judgment. The question arises: how can the classification of spatial objects be improved and the quality of classification assured efficiently in a way that does not require ground truth?
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have