Information Selection Information Selection

This extension includes a set of operators for information selection form the training set for classification and regression problems. These are operators for instance selection (example set selection), instance construction (creation of new examples that represent a set of other instances), clustering, LVQ neural networks, dimensionality reduction, and other. These operators can be used for outlier elimination and training set compression.

Information Selection Extension includes:

Instance selection algorithms such as:

  •     CNN - Condensed Nearest Neighbor Rule
  •     ENN - Edited Nearest Neighbor Rule
  •     RENN - Repeated Edited Nearest Neighbor Rule
  •     All k-NN
  •     RNG - Relative Neighbor Graph
  •     GE - Gabriel Editing
  •     ELH - Encoding Length Heuristic
  •     RMHC – Random Mutation Hill Climbing
  •     IB2
  •     IB3
  •     Random Selection
  •     Weka Drop1-5
  •     Weka ICF
  •     Some other Instance Selection algorithms from Weka

Ensemble Instance Selection algorithms such as:

  •     Ensemble Instance Selection by Bagging
  •     Ensemble Instance Selection by Voting
  •     Ensemble Instance Selection by Feature Subset
  •     Ensemble Instance Selection additional Noise

Generalized Instance Selection

  •     Generalized ENN
  •     Generalized CNN

Competitive based Neural Networks such as:

  •     LVQ1
  •     LVQ2
  •     LVQ2.1
  •     LVQ3
  •     OLVQ
  •     Weighted LVQ
  •     SNG – Supervised Neural Gas
  •     Winer Takes Most LVQ
  •     Generalized LVQ

Clustering algorithms such as:

  •     Fuzzy c-means
  •     Vector Quantization
  •     Conditional Fuzzy c-Means

Feature set reduction algorithms

  •     MDS Multidimensional Scaling  (use external library: Creative Commons License) - this operator is switched off because of incompatible licenses
  •     Feature Selection based on Infosel++ Package (requires external c++ library)

Performance metrics for:

  •     Instance selection
  •     Clustering

Some other useful operators

Product Details

Version 7.0.0
File size 443 kB
Downloads 30428 (2 Today)30428 downloads
Vendor Marcin Blachnik
Category Machine Learning
Released 4/27/16
Last Update 4/27/16 12:00 PM
License AGPL
Product web site
Rating 0.0 stars(0)


Hi. How I Can Use Weka Drop 1-5 in rapidminer?

samfisher593b, 7/25/17 8:12 PM

eouleodes, 11/11/12 10:16 AM

raltersguss, 11/2/12 7:02 PM

raltersguss, 10/27/12 11:58 AM

yremyremy, 10/22/12 8:41 PM
Log in to post comments.