1. |
- Hedefalk, Finn, et al.
(author)
-
Extending the Intermediate Data Structure (IDS) for longitudinal historical databases to include geographic data
- 2014
-
In: Historical Life Course Studies. - 2352-6343. ; 1, s. 27-46
-
Journal article (peer-reviewed)abstract
- The Intermediate Data Structure (IDS) is a standardised database structure for longitudinal historical databases. Such a common structure facilitates data sharing and comparative research. In this study, we propose an extended version of IDS, named IDS-Geo, that also includes geographic data. The geographic data that will be stored in IDS-Geo are primarily buildings and/or property units, and the purpose of these geographic data is mainly to link individuals to places in space. When we want to assign such detailed spatial locations to individuals (in times before there were any detailed house addresses available), we often have to create tailored geographic datasets. In those cases, there are benefits of storing geographic data in the same structure as the demographic data. Moreover, we propose the export of data from IDS-Geo using an eXtensible Markup Language (XML) Schema. IDS-Geo is implemented in a case study using historical property units, for the period 1804 to 1913, stored in a geographically extended version of the Scanian Economic Demographic Database (SEDD). To fit into the IDS-Geo data structure, we included an object lifeline representation of all of the property units (based on the snapshot time representation of single historical maps and poll-tax registers). The case study verifies that the IDS-Geo model is capable of handling geographic data that can be linked to demographic data.
|
|
2. |
|
|
3. |
- Hedefalk, Finn
(author)
-
Life histories across time and space : methods for including geographic factors on the micro-level in longitudinal demographic research
- 2014
-
Licentiate thesis (other academic/artistic)abstract
- Historical demography, which is the study of human population dynamics in the past, is central for understanding human behaviours and traits, such as fertility, mortality and migration. An important factor in demographic research is the geographic context. Where people lived often determined their social ties, exposure to diseases and economic development. Such information is essential not only for historical demographic research but also for a wide range of disciplines.While the geographic context on an aggregated level has an important role in longitudinal historical studies, geographic contexts on a micro-level have only played a minor role. This licentiate contributes to historical demographic research by studying how geographic factors on the micro-level can be included in longitudinal historical analyses. A primary focus is the methodological development for creating longitudinally detailed locations that can be linked to individuals in demographic databases. This research should offer a variety of possibilities for studying how geographic factors on the micro-level affected human living conditions throughout history. The thesis has four research objectives. The first objective is to extend a standardised data model for longitudinal demographic data to include geographic data. This is achieved by introducing IDS-Geo, which is a geographically extended version of the standardised data model IDS. The second objective is to develop and evaluate harmonisation methods to ensure that source data comply with standardised data models. This is achieved by testing and developing a method for first harmonising Swedish environmental data and metadata and then testing the data for compliance against standardised data models and specifications. The third objective is to develop a methodology for creating integrated longitudinal demographic and geographic databases that include geographic factors on the microlevel in demographic research. The core of the methodology is to transform geographic objects in snapshot time representations (digitised from historical maps) into longitudinal object lifeline time representations, and to link individuals to these geographic objects using standardised locations. The methodology is implemented in a case study in which we integrate information from approximately 60 digitised historical maps with longitudinal individual-level data from the Scanian Economic Demographic Database (SEDD). We link 80,431 individuals in five rural parishes in Sweden during 1813-1914 to the property units where they lived. The resulting database is tested using fundamental queries for spatio-temporal data. Additional historical geographic data used for computing context variables are constructed. The results are a unique contribution in terms of linking individuals over such long time periods to longitudinal geographic data on the micro-level. Lastly, the fourth objective of the thesis is to perform longitudinal demographic analyses where geographic factors can subsequently be included. This is performed by analysing the intergenerational effects of child bearing by relatively older women on the longevity of adult offspring in pre-transitional Utah, USA.
|
|
4. |
- Hedefalk, Finn, et al.
(author)
-
Making Swedish environmental geodata INSPIRE compliant : A harmonization case study
- 2011
-
In: Mapping and Image Science. - 1651-8705. ; :3, s. 30-37
-
Journal article (peer-reviewed)abstract
- The European project Nature-SDIplus has developed data and metadata specifications for three INSPIRE Annex III themes: Habitat and Biotopes, Bio-geographical regions and Species distributions. These serves as a foundation for the thematic groups developing the corresponding INSPIRE specifications. The aim of this study is to test a data harmonization approach to make Swedish environmental geodata and metadata compliant with these specifications. In the harmonization process, we use offline transformations that are split into one spatial and one non-spatial part, and standardized formats to allow vendor neutrality. Moreover, we extend the compliance tests to the data and metadata specifications by validating against both extensible Markup Language (XML)-schema and Schematron. Finally, we identify harmonization processes that may be costly or have negative impacts on data quality. The harmonized data and metadata are thereafter published as network services compliant with OGC Web Service specifications. The output from our method is data and metadata that are valid to the Nature-SDIplus data specifications and metadata profiles. Although the usage of standardized formats facilitates vendor neutrality, the nonspatial transformation procedures expressed in interoperable languages seem to be insufficient to execute all the mapping rules. Therefore, some of these transformations cannot be executed in a vendor neutral environment without modifications. Furthermore, by splitting the harmonization into two manageable parts, we avoid some limitations about XML schema translations in existing spatial transformation tools. Additional findings are: (1) by extending the validation with Schematron tests, we find non-compliances that have been missed during the XML schema tests; (2) costly processes are identified, which are caused by missing elements and by unstructured information given as comments; and (3) degradation of the positional and thematic accuracy occur during the harmonization.
|
|
5. |
|
|
6. |
- Rajabi, Mohammadreza, et al.
(author)
-
Comparing Knowledge-Driven and Data-Driven Modeling methods for susceptibility mapping in spatial epidemiology : a case study in Visceral Leishmaniasis
- 2014
-
In: Proceedings of the AGILE'2014 International Conference on Geographic Information Science, Castellón, June, 3-6. ; , s. 1-5
-
Conference paper (peer-reviewed)abstract
- The aim of this study is to compare knowledge-driven and data-driven methods for susceptibility mapping in spatial epidemiology. Our comparison focuses on one of the arguably most important requisites in such models, namely predictability. We compare one data-driven modelling method called Radial Basis Functional Link Net (RBFLN - a well-established Neural Network method) with two knowledge-driven modelling methods, Fuzzy AHP_OWA and Fuzzy GIS-based group decision making (multi criteria decision making methods). These methods are compared in the context of a concrete case study, namely the environmental modelling of Visceral Leishmaniasis (VL) for predictive mapping of risky areas. Our results show that, at least in this particular application, RBFLN model offers the best predictive accuracy
|
|