Geovisualization to support the exploration of large health and demographic survey dataReport as inadecuate

Geovisualization to support the exploration of large health and demographic survey data - Download this document for free, or read online. Document in PDF available to download.

International Journal of Health Geographics

, 3:12

First Online: 04 June 2004Received: 16 March 2004Accepted: 04 June 2004DOI: 10.1186-1476-072X-3-12

Cite this article as: Koua, E.L. & Kraak, MJ. Int J Health Geogr 2004 3: 12. doi:10.1186-1476-072X-3-12


BackgroundSurvey data are increasingly abundant from many international projects and national statistics. They are generally comprehensive and cover local, regional as well as national levels census in many domains including health, demography, human development, and economy. These surveys result in several hundred indicators. Geographical analysis of such large amount of data is often a difficult task and searching for patterns is particularly a difficult challenge. Geovisualization research is increasingly dealing with the exploration of patterns and relationships in such large datasets for understanding underlying geographical processes. One of the attempts has been to use Artificial Neural Networks as a technology especially useful in situations where the numbers are vast and the relationships are often unclear or even hidden.

ResultsWe investigate ways to integrate computational analysis based on a Self-Organizing Map neural network, with visual representations of derived structures and patterns in a framework for exploratory visualization to support visual data mining and knowledge discovery. The framework suggests ways to explore the general structure of the dataset in its multidimensional space in order to provide clues for further exploration of correlations and relationships.

ConclusionIn this paper, the proposed framework is used to explore a demographic and health survey data. Several graphical representations information spaces are used to depict the general structure and clustering of the data and get insight about the relationships among the different variables. Detail exploration of correlations and relationships among the attributes is provided. Results of the analysis are also presented in maps and other graphics.

Electronic supplementary materialThe online version of this article doi:10.1186-1476-072X-3-12 contains supplementary material, which is available to authorized users.

Download fulltext PDF

Author: Etien L Koua - Menno-Jan Kraak


Related documents