PL EN


Preferencje help
Widoczny [Schowaj] Abstrakt
Liczba wyników
2004 | nr 1011 Pozyskiwanie wiedzy i zarządzanie wiedzą | 172--187
Tytuł artykułu

Metody wzorowane na naturze w zadaniach data mining

Warianty tytułu
Nature Based Methods in Data Mining Tasks
Języki publikacji
PL
Abstrakty
Zadanie data mining (drążenie danych) to otrzymywanie użytecznej wiedzy ze znacznej ilości danych. Do głównych zadań data mining należą: klasyfikacja, klasteryzacja (grupowanie) i odkrywanie reguł asocjacji (czyli znajdowanie związków pomiędzy grupami elementów). W artykule przedstawiono zastosowania metod wzorowanych na naturze (algorytmy ewolucyjne i mrówkowe, sieci neuronowe) w zadaniach data mining.
EN
Methods based on Nature are very popular in the field of Artificial Intelligence. Evolutionary Algorithms, Neural Networks and recently, Ant Systems Optimization are used for solving different problems. An advantage of those approaches is ability to produce acceptable solutions in reasonable time. Among others, data mining tasks, such classification, are the subjects of investigations using all mentioned above attempts. The paper presents the experimental study focused on usefulness of Evolutionary Algorithms and Ant System in classification task. The popular test sets (UCI Repository) are used in the presented study. To compare both approaches, the authors propose a set of criteria, such a number of rules, sensitivity, specificity, etc. Obtained results are summarized and discussed. (original abstract)
Słowa kluczowe
Bibliografia
  • Cole R.M.: Clustering with genetic algorithms, 1998, Master Thesis, http://research.cs.uwa.edu.au/reports/ResReport98/ttTheses.
  • DeJong K.A., Spears W.M., Gordon D.F.: Using genetic algorithms for concept learning, "Machine Learning" 13,161-188,1993.
  • Denebourg L., Aron S., Goss S.: The self organizing exploratory pattern of the argentine ant, Journal of Insect Behavior3, 159-168,1990.
  • Diederich A.R., Tickle J.: A Survey and Critique of Techniques for Extracting Rules from Trained Artificial Neural Networks, Knowledge-Based Systems vol. 8, 373-389, 1995.
  • Dorigo, Maniezzo, Colorili: The Ant System: Optimization by a colony of cooperating, IEEE Transactions on Systems, Man, and Cybernetics Part B: Cybernetics (1996) http://citeseer.nj.nec.com/dorigo96ant.html.
  • Flawley, Piatetsky-Shapiro: Knowledge discovery databases: An Overview, AI Magazine, 1991. http://citeseer.nj.nec.com/frawley92knowledge.html.
  • Flockhart I.W., Radcliffe N.J.: A genetic algorithm-based approach to data mining, Proc. 2nd Int. Conf. Knowledge Discovery Data Mining (KDD-96). Portland, OR, Aug. 2-4, 1996, p. 299. http://citeseer.nj.nec.com/context/1783134/44319.
  • Freitas A.: A survey of evolutionary algorithms for data mining and knowledge discovery. To appear (in:) A. Ghosh and S. Tsutsui. (eds.) Advances in Evolutionary Computation. Springer- Verlag, 2002. http://www.ppgia.pucpr.br/~alex/papers.html.
  • Gelati C. D., Kirkpatrick M., Vecchi P.: Optimization by simulated annealing, Science, vol. 220, no. 4598, 671-680, 1983.
  • Hall L.O., Ozyurt I.B., Bezdek J.C.: Clustering with a genetically optimized approach. IEEE Transactions on Evolutionary Computation, 3(2):103-112, July 1999.
  • Janikow C.Z.: A knowledge intensive genetic algorithm for supervised learning, "Machine Learning", 13:189-228. 1993.
  • Kuntz P., Snyers D.: Emergent colonization and graph partitioning, (in:) D. Cliff, P. Husbands, J.A. Meyer, W. Stewart (eds.), Third International Conference on Simulation of Adaptive Behavior. From Animals to Animals, vol. 3, MIT Press, Cambridge, MA,. 494-500. 1994.
  • Kwaśnicka H.: Algorytmy genetyczne w uczeniu SIĘ maszyn. VI konferencja "Pozyskiwanie wiedzy z baz danych", Materiały konferencyjne, Akademia Ekonomiczna, Wrocław, 1998.
  • Lumer E.D, Faieta B.: Diversity and Adaptation in Populations of Clustering Ants. In D. Cliff, P. Husbands, J. Meyer, and S. Wilson (eds.), Procs. of SAB94 - 3 rd Conf. on Simulation of Adaptive Behavior: From Animal to Animals, Cambridge, MA: The MIT Press/Bradford Books. 1994.
  • Ngan Po Shun, Wong Man Leung, Lam Wai: Medical data mining using evolutionary computation. Medical data mining using evolutionary computation. Artificial Intelligence in Medicine 16 (1): 73-96,1999.
  • Nonmarche M., Stimane G. Venturini: AntClass: discovery clusters in numeric data by an hybridization of ant colony with K-means algorithm, 1999. http://citeseer.nj.nec.com/ 439223.html.
  • Park Y-J., Song M-S.: A genetic algorithm for clustering problems, (in:) Proc. of 3rd Annual Conf. on Genetic Programming, 568-575, 1998, http://citeseer.nj.nec. com/context/1799481/0.
  • Parpinelli R.S., Lopes H.S.: An Ant Colony Algorithm for Classification Rule Discovery. http://citeseer.nj.nec.com/458683.html.
  • Pei M., Goodman E.D., Punch W.F.: Pattern discovery from data using genetic algorithms. Proc. 1st Pacific-Asia Conf. Knowledge Discovery & Data Mining (PAKDD-97). Feb. 1997. http://citeseer.nj.nec.eom/context/736965/0.
  • Projekt "MLC++" http://www.sgi.com/tech/mlc.
  • Radcliffe N.J., Surry P.D.: Co-operation through Hierarchical Competition in Genetic Data Mining, Edinburg parallel Computing Centre, Scotland., EPCC-TR94-09: 1994. http://citeseer.nj.nec.com/radcliffe94cooperation.html.
  • Raul Т., Santos J., Nievola C., Freitas A.: Extracting Comprehensible Rules from Neural Networks via Genetic Algorithms. 2000. http://citeseer.nj.nec.com/308761.html.
  • Śmiałek S.: Metody wzorowane na naturze w zadaniach data mining, praca magisterska, Politechnika Wrocławska, 2003.
  • Towell J. Shavlik: Extracting refined rules from knowledge based neural networks. "Machine Learning" 13, 71-101, 1993.
  • Weiss G.M.: Timeweaver: a Genetic Algorithm for Identifying Predictive Patterns in Sequences of Events. Proceedings of the Genetic and Evolutionary Computation Conference 1999. http://citeseer.nj.nec.com/weiss99timeweaver.html.
Typ dokumentu
Bibliografia
Identyfikatory
Identyfikator YADDA
bwmeta1.element.ekon-element-000095657962

Zgłoszenie zostało wysłane

Zgłoszenie zostało wysłane

Musisz być zalogowany aby pisać komentarze.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.