Abstract
The objectives of the study reported in this paper are: (1) to evaluate the adequacy of two data mining techniques, decision tree and neural network in analysing consumer preference for a fast-food franchise and (2) to examine the sufficiency of the criteria selected in understanding this preference. We build decision tree and neural network models to fit data samples collected from 800 respondents in Taiwan to understand the factors that determine their brand preference. Classification rules are generated from these models to differentiate between consumers who prefer the brand and those who do not. The generated rules show that while both decision tree and neural network models can achieve predictive accuracy of more than 80% on the training data samples and more that 70% on the cross-validation data samples, the neural network models compare very favourably to a decision tree model in rule complexity and the numbers of relevant input attributes.
Similar content being viewed by others
References
Agrawal D and Schorling C (1996). Market share forecasting: An empirical comparison of artificial neural networks and multinomial logit model . J Retail 72(4): 383–407.
Bagozzi RP and Warshaw PR (1990). Trying to consume . J Consum Res 17: 127–140.
Bishop CM (1995). Neural Networks for Pattern Recognition . Oxford University Press: New York.
Blin JM and Dodson JA (1980). The relationship between attributes, brand preference, and choice: A stochastic view . Mngt Sci 26(6): 606–619.
Breiman L, Friedman JH, Olshen RA and Stone CJ (1984). Classification and Regression Trees . Wadsworth and Brooks/Cole: San Francisco, CA.
Dennis Jr JE, and Schnabel RE (1983). Numerical Methods for Unconstrained Optimization and Nonlinear Equations . Prentice-Hall: Englewood Cliffs, NJ.
Fish KE, Johnson JD, Dorsey RE and Blodgett JG (2004). Using an artificial neural network trained with a genetic algorithm to model brand share . J Bus Res 57: 79–85.
Fishbein M and Ajzen I (1975). Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research . Addison-Wesley: Reading, MA.
Hornik K, Stinchcombe KM and White H (1989). Multilayer feedforward networks are universal approximators . Neural Networks 2: 359–366.
Hruschka H and Natter M (1999). Comparing performance of feedforward neural nets and K-means for cluster-based market segmentation . Eur J Opl Res 114: 346–353.
Kamakura WA, Kim BD and Lee J (1996). Modeling preference and structural heterogeneity in consumer choice . Market Sci 15(2): 152–172.
Karnani A (1985). Strategic implications of market share attraction models . Mngt Sci 31(5): 536–547.
Laroche M, Hui H and Zhou L (1994). A test of effects of competition on consumer brand selection processes . J Bus Res 31(2/3): 171–181.
Laroche M, Takahashi I, Kalamas M and Teng L (2005). Modeling the selection of fast-food franchises among Japanese consumers . J Bus Res 58(8): 1121–1131.
Liu H and Setiono R (1995). Chi2: a feature selection and discretization of numeric attributes . In: Vassilopoulos JF (ed). Proceedings of the Seventh IEEE International Conference on Tools with Artificial Intelligence, IEEE Computer Society, Press: Los Alamitos, CA, pp. 388–391.
Liu H and Tan ST (1995). X2R: a fast rule generator . In: Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, IEEE: Piscataway, NJ, pp. 1631–1635.
Mitchell T (1997). Machine Learning . McGraw Hill: New York.
Quinlan JR (1993). C4.5 Programs for Machine Learning . Morgan Kaufmann: San Mateo, CA.
Setiono R (1996). A penalty function approach for pruning feedforward neural networks . Neural Comput 7(2): 147–166.
Setiono R and Liu H (1996). Symbolic representation of neural networks . IEEE Comput 29(3): 71–77.
Setiono R, Pan SL, Hsieh MH and Azcarraga A (2005). Automatic knowledge extraction from survey data: learning M-of-N constructs using a hybrid approach . J Opl Res Soc 56(1): 3–14.
Tickle AB, Andrews R, Golea M and Diederich J (1998). The truth will come to light: Directions and challenges in extracting the knowledge embedded within trained artificial neural networks . IEEE Trans Neural Networks 9(6): 1057–1068.
van Wezel M and Potharst R (2007). Improved customer choice predictions using ensemble methods . Eur J Opl Res 181: 436–452.
Vroomen B, Franses PH and van Nierop E (2004). Modeling consideration sets and brand choice using artificial neural networks . Eur J Opl Res 154: 206–217.
Yada K, Ip E and Katoh N (2007). Is this brand ephemeral? A multivariate tree-based decision analysis of new product sustainability . Decis Support Syst 44(1): 223–234.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
The questions included in the questionnaire are as follows.
Group A. Brand attributes: if the respondent agrees that the well-known fast-food franchise possesses the following attributes:
-
1
Clean and tidy
-
2
Fresh ingredients are used
-
3
Easy to find seats/tables
-
4
Serves breakfast items I prefer
-
5
Frequent launch of new products
-
6
Fast and efficient service
-
7
Already had a good experience/impression
-
8
Allows mix-and-match set meal menu items
-
9
A store I feel familiar with
-
10
Serves my favourite food items
-
11
Has TV promotions that attract me
-
12
Offers reasonable price and good value
-
13
Has wide menu variety
-
14
Serves tasty milk tea and coffee
-
15
Suitable to go when in a hurry
-
16
Convenient store locations
-
17
Suitable for the whole family
-
18
Menu consists of breakfast, lunch and dinner
-
19
Most menu offerings suit my taste
-
20
Makes me feel comfortable
-
21
Offer freshly prepared food
-
22
A place that reminds me of my childhood
-
23
Good staff attitude
-
24
Food items are not ‘heaty’
-
25
Offers gift/toy redemption that are attractive
-
26
Serves food with consistent quality
-
27
Offers price discount/coupon
-
28
Food is filling
-
29
Has clean bathrooms
-
30
Is a place adults enjoy
-
31
Serves tasty food
-
32
Is a place children enjoyed
-
33
Offers everyday affordable prices
-
34
Has attractive and comfortable decoration
-
35
Serves burgers I like
-
36
Serves chicken products I like
-
37
Serves French fries I like
-
38
Puts me in a good mood
-
39
Is a part of Taiwan life
-
40
Is an energetic place
-
41
Provides employees with a good working environment
Group B. Perceived importance of certain attributes when deciding whether to patronize a fast-food outlet. The 41 questions in this group are the same as those in Group A.
Group C. Demographic information.
-
1
Whether use internet (Yes/No)
-
2
Frequency of internet use
-
less than once a week,
-
2–6 times a week,
-
than once a day
-
-
3
Marital status
-
Single
-
Partnered/married
-
Separated/divorced
-
Widowed
-
-
4
Number of household members
-
1–3
-
4
-
5
-
6–10
-
-
5
Financial status
-
You are the only one who is financially supporting the household
-
You are not the only one who is financially supporting the household but is the main one
-
You are neither the only one who is financially supporting the household nor is the main one
-
You do not need to financially support the household
-
-
6
Education
-
Primary school or below
-
Junior high school
-
Senior high school/vocational training school
-
University/college
-
Graduate school
-
-
7
Working status
-
Working full-time
-
Working part-time
-
Student
-
Housewife/home-maker
-
Retired
-
Unemployed
-
-
8
Occupation
-
Professional
-
Corporate director/management
-
Business owner/self-employed/merchant
-
Skilled white collar
-
Unskilled white collar
-
Skilled blue collar
-
Unskilled blue collar
-
-
9
Personal monthly income
-
Under NT$20 000
-
NT$20 000–NT$29 999
-
NT$30 000–NT$39 999
-
NT$40 000–NT$59 999
-
NT$60 000 or more
-
-
10
Household monthly income
-
Under NT$40 000
-
NT$40 000–NT$59 999
-
NT$60 000–NT$79 999
-
NT$80 000 or more
-
-
11
Personal eating out expenditure
-
Under NT$1000
-
NT$1000–NT$1999
-
NT$2000–NT$3999
-
NT$4000–NT$5999
-
NT$6000 or more
-
-
12
Household eating out expenditure
-
Under NT$4000
-
NT$4000–NT$5999
-
NT$6000–NT$7999
-
NT$8000 or more
-
-
13
Eating out frequency
-
At least 2 times a day
-
Once a day
-
4–6 times a week
-
2–3 times a week
-
Once a week or less often
-
-
14
Region
-
Greater Taipei
-
Northern Taiwan
-
Central Taiwan
-
Southern Taiwan
-
-
15
Type of region
-
Metropolitan
-
Non-metropolitan
-
-
16
Gender
-
Male
-
Female
-
-
17
Age range
-
15–19
-
20–24
-
24–29
-
30–34
-
35–39
-
40–44
-
45–54
-
Rights and permissions
About this article
Cite this article
Hayashi, Y., Hsieh, MH. & Setiono, R. Predicting consumer preference for fast-food franchises: a data mining approach. J Oper Res Soc 60, 1221–1229 (2009). https://doi.org/10.1057/palgrave.jors.2602646
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1057/palgrave.jors.2602646