#### DMCA

## Interactive Search of Rules in Medical Data Using Multiobjective Evolutionary Algorithms

### Citations

661 | A fast elitist non-dominated sorting genetic algorithm for multi-objective optimization: NSGA-II. Paper presented at the sixth international conference on parallel problem solving from nature, - Deb - 2000 |

179 |
Data mining and knowledge discovery with evolutionary algorithms (Springer,
- Freitas
- 2002
(Show Context)
Citation Context ... the searching process has to besmulticriterial and dozens of such measures have been proposedsand investigated [3, 13].sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining =-=[8]-=- and a significant number of works describe thesuse of EAs in discovering rules from data [2, 7, 11, 12] or in postprocessing the set of rules previously extracted by nonevolutionary approaches [9, 10... |

75 | Fuzzy Rule Selection by MultiObjective Genetic Local Search Algorithms and Rule Evaluation Measures
- Ishibuchi, Yamamoto
- 2004
(Show Context)
Citation Context ...ng [8] and a significant number of works describe thesuse of EAs in discovering rules from data [2, 7, 11, 12] or in postprocessing the set of rules previously extracted by nonevolutionary approaches =-=[9, 10]-=-. Except for the work of Ishibuchisand Yamamoto, they treat the multi-criterial character of thessearch by aggregating all criteria in a single one through a prespecified aggregation function (e.g. a ... |

52 |
Discovering interesting prediction rules with a genetic algorithm,
- Noda, Freitas, et al.
- 1999
(Show Context)
Citation Context ...dsand investigated [3, 13].sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining [8] and a significant number of works describe thesuse of EAs in discovering rules from data =-=[2, 7, 11, 12]-=- or in postprocessing the set of rules previously extracted by nonevolutionary approaches [9, 10]. Except for the work of Ishibuchisand Yamamoto, they treat the multi-criterial character of thessearch... |

43 | Interactive evolutionary multiobjective optimization and decision-making using reference direction method.
- Deb, Kumar
- 2007
(Show Context)
Citation Context ...].s4.1.3 Selection and ArchivingsAfter a new population is created by crossover and mutation, asselection step (typical to MOEAs) is applied. Our selectionsstrategy is similar to that used in NSGA-II =-=[5]-=-, meaning that theselements in the joined population (parents and offsprings) issranked based on the non-domination relationship. A rule issconsidered as non-dominated, with respect to rules in a give... |

40 | Discovering comprehensible classification rules with a genetic algorithm,”
- Fidelis, Lopes, et al.
- 2000
(Show Context)
Citation Context ...dsand investigated [3, 13].sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining [8] and a significant number of works describe thesuse of EAs in discovering rules from data =-=[2, 7, 11, 12]-=- or in postprocessing the set of rules previously extracted by nonevolutionary approaches [9, 10]. Except for the work of Ishibuchisand Yamamoto, they treat the multi-criterial character of thessearch... |

22 | An Evolutionary Algorithm to Discover Numeric Association Rules.
- Mata, Alvarez, et al.
- 2002
(Show Context)
Citation Context ...dsand investigated [3, 13].sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining [8] and a significant number of works describe thesuse of EAs in discovering rules from data =-=[2, 7, 11, 12]-=- or in postprocessing the set of rules previously extracted by nonevolutionary approaches [9, 10]. Except for the work of Ishibuchisand Yamamoto, they treat the multi-criterial character of thessearch... |

20 |
MODENAR: multi-objective differential evolution algorithm formining numeric association rules,”Applied
- Alatas, Akin, et al.
- 2008
(Show Context)
Citation Context ... all criteria in a single one through a prespecified aggregation function (e.g. a product or weighted sum).sApproaches based on multi-objective evolutionary algorithmss(MOEAs) have also been proposed =-=[1, 6]-=-.sHowever, these approaches fail to take into account the user: nosset of quality criteria can be exhaustive, so the user should besinvolved in the search process. In this work, we propose ansapproach... |

14 |
Application of elitist multi-objective genetic algorithm for classification rule generation,”
- Dehuri, Patnaik, et al.
- 2008
(Show Context)
Citation Context ... all criteria in a single one through a prespecified aggregation function (e.g. a product or weighted sum).sApproaches based on multi-objective evolutionary algorithmss(MOEAs) have also been proposed =-=[1, 6]-=-.sHowever, these approaches fail to take into account the user: nosset of quality criteria can be exhaustive, so the user should besinvolved in the search process. In this work, we propose ansapproach... |

8 | Rule discovery with a parallel genetic algorithm,
- Araujo, Lopes, et al.
- 2000
(Show Context)
Citation Context |

8 | Evaluating the correlation between objective rule interestingness measures and real human interest.
- Carvalho, Freitas, et al.
- 2005
(Show Context)
Citation Context ... conflicting, i.e. an accurate rule is not necessarilysinteresting or easy to read, thus the searching process has to besmulticriterial and dozens of such measures have been proposedsand investigated =-=[3, 13]-=-.sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining [8] and a significant number of works describe thesuse of EAs in discovering rules from data [2, 7, 11, 12] or in postp... |

6 | Discovering accurate and interesting classification rules using genetic algorithms. In
- Gopalan, Alhajj, et al.
- 2006
(Show Context)
Citation Context ...ng [8] and a significant number of works describe thesuse of EAs in discovering rules from data [2, 7, 11, 12] or in postprocessing the set of rules previously extracted by nonevolutionary approaches =-=[9, 10]-=-. Except for the work of Ishibuchisand Yamamoto, they treat the multi-criterial character of thessearch by aggregating all criteria in a single one through a prespecified aggregation function (e.g. a ... |

3 | Reconceptualising interestingness metrics for medical data mining.
- Shillabeer, Roddick
- 2005
(Show Context)
Citation Context ...] are:s“IF Plasma glucose concentration in [156,196], Bodysmass index in [8.33,58.64] THEN class=diabetes”s(high accuracy rule)s“IF Plasma glucose concentration in [156,188],sNumber of pregnancies in =-=[3,14]-=-, 2-Hour serum insulins(mu U/ml) in [87,730], Body mass index in [7.9,59.74]sTHEN class=diabetes” (high uncovered negativesmeasure)sTable 3. Measures associated to the set of rules evolved fromsthe Pi... |

1 |
Proposal of medical KDD support user interface utilizing rule interestingness measures.
- Ohsaki, Abe, et al.
- 2006
(Show Context)
Citation Context ... conflicting, i.e. an accurate rule is not necessarilysinteresting or easy to read, thus the searching process has to besmulticriterial and dozens of such measures have been proposedsand investigated =-=[3, 13]-=-.sEvolutionary algorithms (EAs) proved to be valuable instrumentssin data mining [8] and a significant number of works describe thesuse of EAs in discovering rules from data [2, 7, 11, 12] or in postp... |