Differences

This shows you the differences between two versions of the page.

--- protected:limbo [2009/05/12 00:39] – mark
+++ protected:limbo [2010/05/06 14:12] (current) – bil
@@ Line 1: / Line 1: @@
 ====== LIMBO Algorithm ======
 ====== History ======
-Algorithm was developed by Periklis Andritsos and Vassilios Tzerpos.
+The algorithm was developed by Periklis Andritsos and Vassilios Tzerpos.
 ====== Algorithm Intent ======
-Banch algorithm produces [[nested/flat decomposition|nested]]  decomposition from a software system.
+To generate decompositions that exhibit the least information lost when entities are represented by their clusters.
 ====== Factbase Properties ======
-Factbased is generic dependency data table where each row describes one [[terms|entity]] to be cluster. Each column contains the value for a specific [[terms|attribute]].
+The factbase is a generic dependency data table where each row describes one [[terms|entity]] to be clustered. Each column contains the value for a specific [[terms|attribute]].
 ====== Clustering Objectives ======
-The Limbo algorithm produces software decomposition with minimum information loss.
+The Limbo algorithm produces software decompositions with minimum information loss.
 ====== Process Description ======
+An agglomerative clustering algorithm that on each step merges the two clusters with the least information loss.
+The information lost is calculated using the 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 ======
+LIMBO has four phases:
+   - Creation of the Summary Artefacts
+   - Application of the AIB algorithm
+   - Associating original artefacts with clusters
+   - Determining the number of clusters