Uso de minería de datos en el manejo de información geográfica

Abstract

A methodology for the management of data on geographic information is proposed. The methodology was applied to the identification of construction materials of streets and avenues of a defined geographic area in Ciudad Juarez, Mexico. The necessary data were obtained from the Geographic Information System of the Universidad Autónoma de Ciudad Juárez which included the intensity and the coordinates (X, Y, Z) from about 878 thousand records of the city. The development of the method was done in several steps. The first is the pre-processing and cleaning of data; the second step is the model formulation ant its inclusion in a database manager; the third one is the analysis itself with data mining techniques using the WEKA software. The k-means and Make Density Based Clusterer algorithms were utilized for prediction. The results were compared with satellite images of the areas studied, validating in this way the proposed methodology.