ON DESIGN PREFERENCE ELICITATION WITH CROWD IMPLICIT FEEDBACK

We define preference elicitation as an interaction, consisting of a sequence of computer queries and human implicit feedback (binary choices), from which the user’s most preferred design can be elicited. The difficulty of this problem is that, while a humancomputer interaction must be short to be effective, query algorithms usually require lengthy interactions to perform well. We address this problem in two steps. A black-box optimization approach is introduced: The query algorithm retrieves and updates a user preference model during the interaction and creates the next query containing designs that are both likely to be preferred and different from existing ones. Next, a heuristic based on accumulated elicitations from previous users is employed to shorten the current elicitation by querying preferred designs from previous users (the “crowd”) who share similar preferences to the current one.

[1]  Hideyuki Takagi,et al.  Interactive evolutionary computation: fusion of the capabilities of EC optimization and human evaluation , 2001, Proc. IEEE.

[2]  Panos Y. Papalambros,et al.  Use of shape preference information in product design , 2007 .

[3]  D. Dennis,et al.  A statistical method for global optimization , 1992, [Proceedings] 1992 IEEE International Conference on Systems, Man, and Cybernetics.

[4]  Panos Y. Papalambros,et al.  Preference Inconsistency in Multidisciplinary Design Decision Making , 2007, DAC 2007.

[5]  Gediminas Adomavicius,et al.  Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[6]  Risto Miikkulainen,et al.  Evolving Neural Networks through Augmenting Topologies , 2002, Evolutionary Computation.

[7]  D. McFadden.  Conditional logit analysis of qualitative choice behavior , 1972 .

[8]  Donald R. Jones,et al.  Efficient Global Optimization of Expensive Black-Box Functions , 1998, J. Glob. Optim..

[9]  Kenneth O. Stanley,et al.  Picbreeder: A Case Study in Collaborative Evolutionary Exploration of Design Space , 2011, Evolutionary Computation.

[10]  Panos Y. Papalambros,et al.  A Design Preference Elicitation Query as an Optimization Process , 2011 .

[11]  Jeremy J. Michalek,et al.  Linking Marketing and Engineering Product Design Decisions via Analytical Target Cascading , 2005 .

[12]  Wei Chen,et al.  INCORPORATING CUSTOMER PREFERENCES AND MARKET TRENDS IN VEHICLE PACKAGE DESIGN , 2007, DAC 2007.

[13]  Zaïd Harchaoui,et al.  A Machine Learning Approach to Conjoint Analysis , 2004, NIPS.

[14]  Wei Chen,et al.  Enhancing Discrete Choice Demand Modeling for Decision-Based Design , 2003 .

[15]  Bruce E. Ankenman,et al.  Optimal experimental design of human appraisals for modeling consumer preferences in engineering design , 2009, DAC 2008.

[16]  Yi Ren.  Design Preference Elicitation, Identification and Estimation. , 2012 .

[17]  H. Li,et al.  Product Design Selection Under Uncertainty and With Competitive Advantage , 2000 .

[18]  Thorsten Joachims,et al.  Optimizing search engines using clickthrough data , 2002, KDD.

[19]  Brad Johanson,et al.  GP-Music: An Interactive Genetic Programming System for Music Generation with Automated Fitness Raters , 2007 .

[20]  Sung-Bae Cho,et al.  Application of interactive genetic algorithm to fashion design , 2000 .

[21]  T. Evgeniou,et al.  Disjunctions of Conjunctions, Cognitive Simplicity, and Consideration Sets , 2010 .

[22]  Yoav Shoham,et al.  Fab: content-based, collaborative recommendation , 1997, CACM.

[23]  Michael James Sasena,et al.  Flexibility and efficiency enhancements for constrained global design optimization with kriging approximations. , 2002 .

[24]  Peter J. Fleming,et al.  The MATLAB genetic algorithm toolbox , 1995 .

[25]  A. Lloyd.  Threats to the estimation of benefit: are preference elicitation methods accurate? , 2003, Health economics.

[26]  Eric T. Bradlow,et al.  Beyond conjoint analysis: Advances in preference measurement , 2008 .