Geospatial Data Mining

Geospatial Data Mining is a set of facilities that aim to (i) compare two geospatial distributions, (ii) to retrieve spatiotemporal information from a remotely hosted geospatial layer and (iii) to perform data mining on geographical layers containing environmental information. Geospatial data mining is included in the EcologicalEngineGeoSpatialExtension library of the gCube framework, as it relies on the data mining processes contained in the EcologicalEngine library.

Overview

Geospatial processing is useful in many applications of marine sciences, including Niche Modeling, Vessels Information processing, Ecological Modeling and Biodiversity monitoring. Environmental characteristics are usually put in the format of n-dimensional vector of real values. Such vectors must be as independent as possible in order to properly describe a phenomenon. Dependent vectors correspond to redundant information are not useful to automatic models. Geospatial processing includes procedures to retrieve environmental information in the format of n-dimensional vectors and the processing needed to evaluate the differences between two datasets or the degree of completeness of a single dataset. This can be essential in calculating the difference between the presence distributions of a certain species in two different years, in order to understand if the distribution is wider or narrower.

Features

The features currently supported by the Geospatial data mining facilities include:

• Environmental layers indexing on a GeoNetwork instance, with respect to the ISO19115:2003 specifications.
• Retrieval of environmental parameters information associated to a coordinates triple. Such information is given according to the time instants included in the layer.
• Retrieval of environmental parameters values associates to a set of points and an a time instant.
• Automatic simulation of values in the points in which information is not defined.
• Management of WFS and OpenDap based layers in seamless way to the library users.

Software

The software is available on the gCube maven repository by including the following component in the pom.xml file:

<dependency>
  <groupId>org.gcube.dataanalysis</groupId>
  <artifactId>ecological-engine-geospatial-extensions</artifactId>
  <version>1.0.0-SNAPSHOT</version>
</dependency>

An example to call the spectrogram analysis with STFT and produce the chart is:

SignalConversions.spectrogram(name, signal, samplingRate, windowshift, frameslength, display)

Where the input variables are:

String name: the title of the chart
double[] signal: the sequence of values representing the trend
int samplingRate: the sampling frequency in integer value and multiple of 2
int windowshift: the window shift of the STFT in samples
int frameslength: the length of each window in samples
boolean display: a flag to ask the procedure to run an applet which displays the spectrogram

An example which performs a signal reconstruction is:

AlgorithmConfiguration config = new AlgorithmConfiguration();
config.setConfigPath(configDir);
config.initRapidMiner();
SignalProcessing.fillSignal(signal)

where the input parameters are defined as follows:

double[] signal: the sequence of values representing the trend
String configDir: a configuration folder containing the configuration files required by the Ecological Engine library

The cfg directory and the Ecological Engine library are accessible at this svn link: http://svn.research-infrastructures.eu/d4science/gcube/trunk/data-analysis/EcologicalEngine