Interactive genetic algorithms ( iGAs )are 皿 ethods that acquire and analyse nser preference based on user , s sub − jective evaluation . IGA has been applied to various unimedal problems , such as parameter setting of a hear 主 ng aid and fashiDn des } gn . On the other hand, the goal of this study is achieving iGA which also corresponds to multimoda 玉 preferences wj もh equivalent 且tness at the peaks . R 〕r example , when users select products on shopplng sites , they have several types of

more »

... rence trends at the same time . In this case , renecting all the trends in product presentation leads to increased sales and consumer satisfaction . The dependency among desig囂 variable8 should be also considercd toward each trend ef preference . In this study , a new of 聴 pring generation method that enables ef 丘cient search even if user preference is multimodal and there are dependencies among design variables is proposed . The proposed method is capable to detect a mu Σ timodal pre 魚 rence using clustering , and spawn off − spring based on probabilistic rnodel eliminating the dependencies among design variables by principal component analysis . The results of experiments indicate that the proposed method can generate offspring corresponding to multjmodal preferences with equiva 至 ent 且tness at the peaks and dependencies among design variables . However , there are f 合 w subjects who have dependencies within their pref α ences . Key word : interactive genetic algorithms , offspring generation, multimodality , dependency