GrowingSOM (version 0.1.1)

GrowingSOM: \Sexpr[results=rd,stage=build]{tools:::Rd_package_title("#1")}GrowingSOMGrowing Self-Organizing Maps

Description

\Sexpr[results=rd,stage=build]{tools:::Rd_package_description("#1")}GrowingSOMA growing self-organizing map (GrowingSOM, GSOM) is a growing variant of the popular self-organizing map (SOM). A growing self-organizing map is a type of artificial neural network (ANN) that is trained using unsupervised learning to produce a two-dimensional representation of the input space of the training samples, called a map.

Details

The GSOM was developed to address the issue of identifying a suitable map size in the SOM. It starts with a minimal number of nodes and grows new nodes on the boundary based on a heuristic. By using the value called Spread Factor (SF), the growth of the GSOM can be controlled. This R-Package supports both a supervised and an unsupervised version of Growing Self-Organizing maps. Maps can be trained, new data can be mapped to existing maps and in the case of supervised GSOM maps, the dependant values can be predicted. \Sexpr[results=rd,stage=build]{tools:::Rd_package_indices("#1")}GrowingSOM Index: This package was not yet installed at build time.

References

This package is conceptually mostly based on the version of Growing Self-Organizing Maps as described in this paper: Damminda Alahakoon, Saman K. Halgamuge (2000): Dynamic Self-Organizing Maps with Controlled Growth for Knowledge Discovery. IEEE TRANSACTIONS ON NEURAL NETWORKS, VOL. 11.

Examples

  ################
  # Example 1
  ################

  # Get some data
  data("iris")
  s = sample(1:150, 100)
  train_set = iris[s,1:4]
  test_set = iris[-s,1:4]
  
  # Train GrowingSOM
  gsom_map <- train.gsom(train_set, spreadFactor=0.8)
  
  # Some Plots
  plot(gsom_map, type = "training")
  plot(gsom_map, type="count")
  par(mfrow=c(2,2))
  plot(gsom_map, type="property")
  par(mfrow=c(1,1))
  
  # Map new data
  gsom_mapped = map.gsom(gsom_map, test_set)
  plot(gsom_mapped)
  
  ################
  # Example 2
  ################
  
  # load data
  data("auto_mpg")
  s = sample(1:392, 300)
  train_set = auto_mpg[s,1:8]
  test_set = auto_mpg[-s,1:8]
  
  # Train Gsom Model (hexagonal grid)
  gsom_map <- train_xy.gsom(train_set[,2:8], train_set[,1], spreadFactor = 0.9, nhood="hex")
  
  print(gsom_map)
  plot(gsom_map, type = "predict")
  
  # Predict mpg for the test set
  gsom_predictions = predict.gsom(gsom_map, test_set[,2:8])