"Madeira" . . "Madeira" . . "978-989-8111-35-7" . . "INSTICC" . "Slu\u010Dov\u00E1n\u00ED rad z v\u00EDce poradn\u00EDch syst\u00E9m\u016F - s aplikac\u00ED na datech z v\u00E1lcovac\u00ED stolice"@cs . "P(1ET100750401), P(1M0572), Z(AV0Z10750506)" . "2008-05-11+02:00"^^ . . . "18 variant poradn\u00EDho syst\u00E9mu bylo vyvinuto na z\u00E1klad\u011B r\u016Fzn\u00FDch pr\u011Bdpoklad\u016F. Proto\u017Ee tyto p\u0159edpokaldy mohou b\u00FDt r\u016Fzn\u011B vhodn\u00E9 pro r\u016Fzn\u00E9 re\u017Eimy \u0159\u00EDzen\u00E9ho syst\u00E9mu, byl vyvinut automatick\u00FD syst\u00E9m slu\u010Dov\u00E1n\u00ED p\u0159edpov\u011Bd\u00ED generovan\u00FDch p\u016Fvodn\u00EDmi syst\u00E9my. P\u0159\u00EDstup byl testov\u00E1n na datech nasb\u00EDran\u00FDch na poradn\u00EDm syst\u00E9mu v\u00E1lcovac\u00ED stolice za dobu 6 m\u011Bs\u00EDc\u016F. Proto\u017Ee oper\u00E1to\u0159i nemuseli uposlechnout p\u0159edkl\u00E1dan\u00E9 rady, bylo vyhodnocen\u00ED kvality porad provedeno na z\u00E1klad\u011B dvou krit\u00E9ri\u00ED: kvalita oper\u00E1torsk\u00FDch z\u00E1sah\u016F a shoda porad se skute\u010Dn\u011B proveden\u00FDm z\u00E1sahem. Rady z navr\u017Een\u00E9ho slu\u010Dov\u00E1n\u00ED byly shled\u00E1ny kvalitn\u011Bj\u0161\u00EDmi ne\u017E rady z kter\u00E9hokoliv p\u016Fvodn\u00EDho syst\u00E9mu."@cs . "Slu\u010Dov\u00E1n\u00ED rad z v\u00EDce poradn\u00EDch syst\u00E9m\u016F - s aplikac\u00ED na datech z v\u00E1lcovac\u00ED stolice"@cs . . . . "RIV/67985556:_____/08:00324357!RIV09-MSM-67985556" . "3"^^ . . "Proceedings of the Fifth International Conference on Informatics in Control, Automation and Robotics" . "4"^^ . . . "\u0160m\u00EDdl, V\u00E1clav" . "378573" . "K\u00E1rn\u00FD, Miroslav" . "Ettler, P." . "Previously, 18 different variants of an advisory system has been developed using different assumptions. Since these assumption may be valid only in certain regimes, we design an automatic procedure for merging of their advices. The approach was tested on data collected on an advisory system of a rolling mill during 6month of operation. Since the operators did not have to follow the advices, the performance of the advisory system is assessed from two measures: operators' performance index and coincidence of the advised and performed action. It was shown that the merged advisory system outperform any of the original systems." . . "Previously, 18 different variants of an advisory system has been developed using different assumptions. Since these assumption may be valid only in certain regimes, we design an automatic procedure for merging of their advices. The approach was tested on data collected on an advisory system of a rolling mill during 6month of operation. Since the operators did not have to follow the advices, the performance of the advisory system is assessed from two measures: operators' performance index and coincidence of the advised and performed action. It was shown that the merged advisory system outperform any of the original systems."@en . . "[F639B776EF80]" . "Merging of advices from multiple advisory systems - with evaluation on rolling mill data"@en . . . "Merging of advices from multiple advisory systems - with evaluation on rolling mill data"@en . "RIV/67985556:_____/08:00324357" . . . . . "Andr\u00FDsek, Josef" . "6"^^ . "Merging of advices from multiple advisory systems - with evaluation on rolling mill data" . "advisory system; Bayesian decision making; Bayeisan model averaging; multiple-participant decision-making"@en . "Merging of advices from multiple advisory systems - with evaluation on rolling mill data" . . . .