Evolutionary and metaheuristics based data mining

DM poses a great range of interesting (NP-hard) problems that consists mainly in searching for: the pattern/model that best describe the data, the more predictive subset of variables, the more accurate parameter configuration, etc. Because of this, and of the success obtained by metaheuristics when...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Soft computing (Berlin, Germany) Ročník 13; číslo 3; s. 209 - 212
Hlavní autori: del Jesús, María J., Gámez, José A., Puerta, José M.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Berlin/Heidelberg Springer-Verlag 01.02.2009
Springer Nature B.V
Predmet:
Artificial Intelligence
Bayesian analysis
Classification
Classifiers
Clustering
Combinatorial analysis
Computational Intelligence
Control
Data mining
Dependence
Editorial
Engineering
Evolutionary algorithms
Fuzzy sets
Genetic algorithms
Heuristic methods
Machine learning
Mathematical Logic and Foundations
Mechatronics
Neural networks
Optimization
Robotics
Variables
ISSN:1432-7643, 1433-7479
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:DM poses a great range of interesting (NP-hard) problems that consists mainly in searching for: the pattern/model that best describe the data, the more predictive subset of variables, the more accurate parameter configuration, etc. Because of this, and of the success obtained by metaheuristics when applied to other combinatorial/numerical optimization problems, metaheuristics have been widely applied to solve DM problems during the last years. [...]the use of evolutionary algorithms and metaheuristics in general to approach DM-based problems (EMBDM) is a hot topic of research nowadays. [...]there is a very active related field of research known as learning classifier systems (LCS) (Bull et al. 2002) that learns the rule-based classifier by combining evolutionary algorithms with reinforcement learning. In descriptive data mining there is no class variable so we are in an unsupervised setting. [...]the goal now is not to predict a target value, but to discover intrinsic relations, dependences, etc. of the data. On the contrary of Bayesian networks that represent a global (in)dependence model among all the variables, association rules can be viewed as a set of separate dependence sentences, where each rule stands for the dependence relation between the variables in the antecedent and the one in the consequent. Because different measures (e.g. support, confidence, interestingness) are used to evaluate a the goodness of a rule, the multiobjective approach has shown to be a good tool to deal with it (Ghosh and Nath 2004).
Bibliografia:SourceType-Scholarly Journals-1
content type line 14
ObjectType-Editorial-2
ObjectType-Commentary-1
ISSN:1432-7643
1433-7479
DOI:10.1007/s00500-008-0373-1