Zusammenfassung
In aktuellen Arbeiten zu Produktempfehlungssystemen wird die wahlbasierte Conjoint-Analyse zur Messung von Benutzerpräferenzen vorgeschlagen. Diese Methode erzielt eine hohe Empfehlungsqualität und leidet nicht unter dem Start-up-Problem, weil sie auch für neue Nutzer und neue Produkte Empfehlungen generiert. Die Anwendung der wahlbasierten Conjoint- Analyse bedeutet für Konsumenten jedoch einen erheblichen Aufwand, der zu einer Abneigung gegenüber derartigen Empfehlungssystemen führt. In diesem Artikel werden mit einer Simulation die hohe Entscheidungsqualität und der hohe Benutzeraufwand eines nutzenbasierten Systems mit wahlbasierten Conjoint-Analysen mit hierarchischem Bayes’-Schätzer aufgezeigt. Um den Widerspruch zwischen hoher Empfehlungsgüte und niedrigem Aufwand aufzulösen wird ein neuer Ansatz entwickelt, der nur Pareto-effiziente Alternativen zeigt und diese anhand der Anzahl der dominierten Attribute sortiert. Es zeigt sich, dass diese rangbasierte Pareto-Front zu einer besseren Empfehlungsliste führt als die Anwendung der wahlbasierten Conjoint-Analyse. Zudem ist der Aufwand für Konsumenten sehr gering und vergleichbar mit sehr einfachen Sortierverfahren.
Abstract
In recent studies on recommendation systems, the choice-based conjoint analysis has been suggested as a method for measuring consumer preferences. This approach achieves high recommendation accuracy and does not suffer from the start-up problem because it is also applicable for recommendations for new consumers or of new products. However, this method requires massive consumer input, which causes consumer reluctance. In a simulation study, we demonstrate the high accuracy, but also the high user’s effort for using a utility-based recommendation system using a choice-based conjoint analysis with hierarchical Bayes estimation. In order to reduce the conflict between consumer effort and recommendation accuracy, we develop a novel approach that only shows Pareto-efficient alternatives and ranks them according to the number of dominated attributes. We demonstrate that, in terms of the decision accuracy of the recommended products, the ranked Pareto-front approach performs better than a recommendation system that employs choice-based conjoint analysis. Furthermore, the consumer’s effort is kept low and comparable to that of simple systems that require little consumer input.
Notes
Diese Ansätze benötigen Daten anderer Konsumenten in der gleichen Produktkategorie (100 Konsumenten in der gleichen Kategorie sind zumeist ausreichend).
Für eine Einführung in die multiattributive Nutzentheorie siehe Wallenius et al. (2008).
Ein D-optimales Design wird verwendet, um die Varianz der Beobachtungsvariable (hier die Produktauswahl) und die Anzahl an Stimuli, die notwendig sind, um die Haupteffekte (hier die Attribute) zu untersuchen, zu minimieren.
Zum Vergleich der Güte und des Aufwands unserer Systeme berechnen wir ANOVAs mit Tukeys Post-hoc-Test mit einem Signifikanzniveau von 5 %.
Das Sortieren von α Attributausprägungen kostet αld(α) READ- und COMPARE-Operationen.
Literatur
Adomavicius G, Tuzhilin A (2005) Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions. IEEE Trans Knowl Data Eng 17(6):734–749
Aksoy L, Cooil B, Lurie NH (2011) Decision quality measures in recommendation agents research. J Interact Mark 25(2):110–122
Ansari A, Essegaier S, Kohli R (2000) Internet recommendation systems. J Mark Res 37(3):363–375
Bettman JR, Johnson EJ, Payne JW (1990) A componential analysis of cognitive effort in choice. Organ Behav Hum Decis Process 45(1):111–129
Biggs SF, Bedard JC, Gaber BG, Linsmeier TJ (1985) The effect of task size and similarity on the decision behavior of bank loan officers. Manage Sci 31(8):970–987
Bröder A, Schiffer S (2003) Bayesian strategy assessment in multi-attribute decision research. J Behav Decis Mak 16(3):193–213
Burke R (2002) Hybrid recommender systems: survey and experiments. User Model User-Adapt Interact 12(4):331–370
Butler JC, Dyer JS, Jia J, Tomak K (2008) Enabling e-transactions with multi-attribute preference models. Eur J Oper Res 186(2):748–765
Butler JC, Morrice DJ, Mullarkey PW (2001) A multiple attribute utility theory approach to ranking and selection. Manage Sci 47(6):800–816
Cao Y, Li Y (2007) An intelligent fuzzy-based recommendation system for consumer electronic products. Expert Syst Appl 33(1):230–240
De Bruyn A, Liechty JC, Huizingh EK, Lilien GL (2008) Offering online recommendations with minimum customer input through conjoint-based decision aids. Mark Sci 27(3):443–460
Denstadli JM, Lines R (2007) Conjoint respondents as adaptive decision makers. Int J Mark Res 49(1):117–132
Dorner V, Scholz M (2013) Predicting and economically exploiting utility thresholds with utility-based recommendation systems. In: Proc of the 21st eur conf inform syst
Fiebig DG, Keane MP, Louviere J, Wasi N (2010) The generalized multinomial logit model: accounting for scale and coefficient heterogeneity. Mark Sci 29(3):393–421
Fritz M, Schlereth C, Figge S (2011) Empirical evaluation of fair use flat rate strategies for mobile Internet. Bus Inf Syst Eng 3(5):269–277
Gensler S, Hinz O, Skiera B, Theysohn S (2012) Willingness-to-pay estimation with choice-based conjoint analysis: addressing extreme response behavior with individually adapted designs. Eur J Oper Res 219(2):368–378
Godfrey P, Shipley R, Gryz J (2006) Algorithms and analyses for maximal vector computation. VLDB J 16(1):5–28
Green PE, Krieger AM, Winder Y (2001) Thirty years of conjoint analysis: reflections and prospects. Interfaces 31(3):56–73
Häubl G, Trifts V (2000) Consumer decision making in online shopping environments: the effects of interactive decision aids. Mark Sci 19(1):4–21
Hevner AR, March ST, Park J, Ram S (2004) Design science in information systems research. MIS Q 28(1):75–105
Hey JD (1982) Search for rules for search. J Econ Behav Organ 3:65–81
Hinz O, Eckert J (2010) The impact of search and recommendation systems on sales in electronic commerce. Bus Inf Syst Eng 2(2):66–77
Hostler RE, Yoon VY, Guo Z, Guimaraes T, Forgionne G (2011) Assessing the impact of recommender agents on on-line consumer unplanned purchase behavior. Inf Manag 48(8):336–343
Hoyer WD (1984) An examination of consumer decision making for a common repeat purchase product. J Consum Res 11(3):822–829
Huang S (2011) Designing utility-based recommender systems for e-commerce: evaluation of preference-elicitation methods. Electron Commer Res Appl 10(4):398–407
Jedidi K, Kohli R, DeSarbo WS (1996) Consideration sets in conjoint analysis. J Mark Res 33(3):364–372
Johnson EJ, Payne JW (1985) Effort and accuracy in choice. Manage Sci 31(4):395–414
Kamis A, Koufaris M, Stern T (2008) Using an attribute-based decision support system for user-customized products online: an experimental investigation. MIS Q 32(1):159–177
Kanninen BJ (2002) Optimal design for multinomial choice experiments. J Mark Res 39(2):214–227
Kumar N, Benbasat I (2006) The influence of recommendations and consumer reviews on evaluations of websites. Inf Syst Res 17(4):425–439
Lohse GR, Johnson EJ (1996) A comparison of two process tracing methods for choice tasks. Organ Behav Hum Decis Process 68(1):28–43
Loomes G, Sugden R (1982) Regret theory: an alternative theory of rational choice under uncertainty. Econ J 92(368):805–824
Montgomery A, Hosanagar K, Krishnan R, Clay K (2004) Designing a better shopbot. Manage Sci 50:189–206
Moore WL (2004) A cross-validity comparison of rating-based and choice-based conjoint analysis models. Int J Res Mark 21(3):299–312
Moore WL, Gray-Lee J, Louviere JJ (1998) A cross-validity comparison of conjoint analysis and choice models at different levels of aggregation. Mark Lett 9(2):195–207
Olshavsky RW (1979) Task complexity and contingent processing in decision making: a replication and extension. Organ Behav Hum Perform 24:300–316
Payne JW, Bettman JR, Johnson EJ (1993) The adaptive decision maker. Cambridge University Press, Cambridge
Pfeiffer J (2012) Interactive decision aids in e-commerce. Springer, Berlin
Rand W, Rust RT (2011) Agent-based modeling in marketing: guidelines for rigor. Int J Mark Res 28(3):181–193
Rieskamp J, Hoffrage U (2008) Inferences under time pressure: how opportunity costs affect strategy selection. Acta Psychol 127(2):258–276
Rossi PE, Allenby GM, McCulloch R (2005) Bayesian statistics and marketing. Wiley, West Sussex
Schlereth C, Eckert C, Skiera B (2012) Using discrete choice experiments to estimate willingness-to-pay intervales. Mark Lett 23(3):761–776
Scholz M, Dorner V (2012) Estimating optimal recommendation set sizes for individual consumers. In: Proc of the 33rd int conf inform syst
Svenson O (1979) Process descriptions of decision making. Organ Behav Hum Perform 23:86–112
Todd PA, Benbasat I (1992) An experimental investigation of the impact of computer based decision aids on processing effort. MIS Q 16(3):373–393
Todd PA, Benbasat I (1994) The influence of decision aids on choice strategies and conditions of high cognitive load. IEEE Trans Syst Man Cybern 24(4):537–547
van Ittersum K, Pennings J (2012) Attribute-value functions as global interpretations of attribute importance. Organ Behav Hum Decis Process 119(1):89–102
Wallenius J, Dyer JS, Fishburn PC, Steuer RE, Zionts S, Deb K (2008) Multiple criteria decision making, multiattribute utility theory: recent accomplishments and what lies ahead. Manage Sci 54(7):1336–1349
Wang W, Benbasat I (2009) Interactive decision aids for consumer decision making in e-commerce: the influence of perceived strategy restrictiveness. MIS Q 33(2):293–320
Wohlin C, Runeson P, Höst M, Ohlsson MC, Regnell B, Wesslén A (2012) Experimentation in software engineering. Springer, Berlin
Xiao B, Benbasat I (2007) E-commerce product recommendation agents: use, characteristics, and impact. MIS Q 31(1):137–209
Yee M, Dahan E, Hauser JR, Orlin J (2007) Greedoid-based noncompensatory inference. Mark Sci 26(4):532–549
Author information
Authors and Affiliations
Corresponding author
Additional information
Angenommen nach einer Überarbeitung durch Prof. Dr. Hinz.
This article is also available in English via http://www.springerlink.com and http://www.bise-journal.org: Pfeiffer J, Scholz M (2013) A Low-Effort Recommendation System with High Accuracy. A New Approach with Ranked Pareto-Fronts. Bus Inf Syst Eng. doi: 10.1007/s12599-013-0295-z.
Zusätzliche Information
Ergänzendes Onlinematerial kann unter folgendem Link abgerufen werden.
Rights and permissions
About this article
Cite this article
Pfeiffer, J., Scholz, M. Produktempfehlungssysteme mit minimalem Konsumentenaufwand und hoher Genauigkeit. Wirtschaftsinf 55, 395–408 (2013). https://doi.org/10.1007/s11576-013-0388-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11576-013-0388-9