Exploring the possibilities of production of a culture codes dataset on cadastral parcel level by making use of external source datasets

Exploring the possibilities of production of a culture codes dataset on cadastral parcel level by making use of external source datasets

Dopper, E.

Stoter, J. (mentor)
Van Buren, J. (mentor)
Leenders, G. (mentor)
Van Oosterom, P. (mentor)

OTB Research Institute for the Built Environment

GIS technology

GIMA

2011-11-01

Sharing and reusing spatial data becomes more and more important. One way of reusing shared spatial data is by creating a new dataset by making use of spatial data available in a combination of external datasets. This has been the central theme in this thesis, as its main objective has been to explore the possibilities of producing a culture codes dataset by making use of external source datasets. Furthermore, the quality of these source datasets, as well as their semantic, geometric and temporal characteristics have been taken into account. The first step has been to describe and analyze a number of candidate source datasets, in order to be able to make a decision on which source datasets to use for the production of the culture codes. This included visual mappings between the classes and attributes of the candidate source datasets and the culture codes. Four source datasets have been selected (AKR, LKI, BAG and TOP10NL), with the choice for these datasets being mainly based on their contents, quality and accessibility. After selecting these source datasets the visual mappings were translated into seventy-one decision rules for the production of the culture codes. Next, these decision rules were integrated into an extensive decision tree. The decision rules and the decision tree have been tested in a proof of concept. This proof of concept has been carried out in four case studies and quality analyses. The results of this proof of concept have been quite satisfying, with 80,3% of all results being right. In order to further improve the results of the production method, a number of additional datasets has to be used and a few decision rules need to be investigated. Furthermore, the large amount of spatial data used for the calculations has caused serious performance problems. This should get special attention when deciding to apply an (automated) production of the culture codes by making use of the proposed production method. However, in general, the thesis shows that producing the culture codes dataset by making use of external datasets should be seen as a serious option.

http://resolver.tudelft.nl/uuid:45864885-760e-4941-8708-335430fc2577

Student theses

master thesis

(c) 2012 Dopper, E.