User Tools

Site Tools


protected:limbo

This is an old revision of the document!


LIMBO Algorithm

History

Algorithm was developed by Periklis Andritsos and Vassilios Tzerpos.

Algorithm Intent

Banch algorithm produces flat decomposition from a software system.

Factbase Properties

Factbased is generic dependency data table where each row describes one entity to be cluster. Each column contains the value for a specific attribute.

Clustering Objectives

The Limbo algorithm produces software decomposition with minimum information loss.

Process Description

An agglomerative clustering algorithm that on each step merges two clusters with least information loss.

The information lost calculates using formulas from Agglomerative Information Bottleneck algorithm.

Decomposition Properties

The LIMBO decomposition has a small value for its information lost function.

Algorithm Restrictions

Failed Assumptions

Detailed Algorithm Description

The LIMBO algorithm has four phases:

  1. Creation of the Summary Artefacts
  2. Application of the AIB algorithm
  3. Associating original artefacts with clusters
  4. Determining the number of clusters
protected/limbo.1242099886.txt.gz · Last modified: 2009/05/12 03:44 by mark