[Solomonov Seminar] 134. Solomonov seminar

Marko Grobelnik marko.grobelnik at ijs.si
Mon Dec 1 13:28:39 CET 2003

Vabim vas na 134. Solomonov seminar, ki bo v torek,
2. decembra 2003 ob 13:00 uri v Veliki predavalnici IJS.
Posnetki in materiali preteklih seminarjev so dostopni
na http://solomon.ijs.si

Na tokratnem seminarju bo gost odseka, Thiemo Krink iz Danske
predstavil nekaj zanimivih alternativnih pristopov k razvrscanju v
skupine (clustering) - to podrocje, je s pojavom data-mininga dozivelo
precejsen razcvet, predvsem pri resevanju problema v zahtevnih
realisticnih situacijah.

Thiemo Krink, University of Aarhus, Denmark

Differential Evolution and Particle Swarm Optimization in Partitional Clustering

In recent years, many partitional clustering algorithms based on genetic algorithms (GA) have been proposed to tackle the problem of
finding the optimal partition of a data set. Surprisingly, very few studies considered alternative stochastic search heuristics
other than GAs or simulated annealing. Two promising algorithms for numerical optimization, which are hardly known outside the
heuristic search field, are particle swarm optimisation (PSO) and differential evolution (DE). In this study, we compared the
performance of GAs with PSO and DE for a medoid evolution approach to clustering. Moreover, we compared these results with the
nominal classification, k-means and random search (RS) as a lower bound. Our results show that DE is clearly and consistently
superior compared to GAs and PSO for hard clustering problems, both in respect to precision as well as robustness (reproducibility)
of the results. Only for trivial problems all algorithms can obtain comparable results. Apart from superior performance, DE is very
easy to implement and requires hardly any parameter tuning compared to substantial tuning for GAs and PSOs. Our study shows that DE
rather than GAs should receive primary attention in partitional cluster algorithms.

More information about the Solomonov-seminar mailing list