Preference mapping

The preferred map scales and data densities for delineation of different scale geochemical patterns are as follows. 

Preference mapping can be accomplished with projection techniques such as multidimensional scaling and cluster analysis, but the following discussion focuses on principal components analysis (PCA) [69] because of the interpretability of the results. A PCA represents a multivariate data table, e.g., N rows ( molecules ) and K columns ( properties ), as a projection onto a low-dimensional table so that the original information is condensed into usually 2-5 dimensions. The principal components scores are calculated by forming linear combinations of the original variables (i.e., properties ). These are the coordinates of the objects ( molecules ) in the new low-dimensional model plane (or hyperplane) and reveal groups of similar... 

Figure 12.1 Multivariate preference mapping of a set of 1064 compounds that were profiled with 11 descriptors (see text for explanation). The upper scores plot shows the projection of the molecules onto the first two components of the principal components analysis. The lower loadings plot illustrates how the descriptors contribute to the positions of the projected compounds in the scores plot.

Eventually, it should be stressed that rapid does not necessarily mean instantaneous. Rather, one should consider rapid proflling methods as ways to acquire data more rapidly than with their conventional equivalents. In Chapter 20, Blumenthal and Herbeth report that in order to evaluate sensations while driving, each assessor in their study drove a total of more than 25 h and 600 km, even if they used a quicker approach than conventional proflling. The same goes with other methods TDS is not as rapid, but it is much faster than traditional time-intensity measurements the Ideal Profile Method may also take some time for the panellists, but it is clearly more rapid than a full external preference mapping study. 

Eventually, Blumenthal et al. (2000) have argued that when experts differ in their description of the product set, it is wiser to look at their sensory maps separately. In such cases, they showed that individual sensory maps could be used selectively in order to improve the goodness of fit in preference mapping smdies. 

Blumenthal, D., Dairou, V, Sieffermann, J. M. and Danzart, M. (2000). How to improve sensory information provided by free choice profiling in preference mapping using individual maps The 5th Sensometrics Meeting. Columbia, Missouri. 

Faye, R, Bremaud, D., Teillet, E., Courcoux, R, Giboreau, A. and Nicod, H. (2006). An alternative to external preference mapping based on consumer perceptive mapping. Food Quality... 

CATA qnestions can be for performing internal preference mapping by projecting consumers desalplions into the preference space. Besides, Ares et al. (2010a) and Dooley et al. (2010) nsed CATA qnestions to generate a consumer-based sensory space for external preference mapping. In these situations, consumers should evaluate at least six samples (Lawless and Heymann, 2010). 

Dooley L., Lee, Y.S. and Meullenet, J.F. (2010), The application of check-all-that-apply (CATA) consumer profiling to preference mapping of vanilla ice cream and its comparison to classical external preference mapping, Food Qual Prefer, 21, 394 01. 

Parente, M.E., Manzoni, A.V. and Ares, G. (2011), External preference mapping of commercial antiaging creams based on consumers responses to a check-all-that-apply question, J Sensory Stud, 26, 158-166. 

It provides similar results when compared to classical preference mapping or profiling procedures. 

Ares, G., Gimenez, A., Barreiro, C. and Gambaro, A. (2010) Use of an open-ended qnestion to identify drivers of liking of milk desserts. Comparison with preference mapping tech-niqnes, Food Qual Prefer, 21, 286-294. DOI 10.1016/j.foodqnal.2009.05.006. 

Elmore, J. R., Heymann, H., Johnson, J. and Hewett, J. E. (1999) Preference mapping Relating acceptance of creaminess to a descriptive sensory map of a semi-solid, Food Qual Prefer, 10, 465-175. DOI 10.1016/S0950-3293(99)00046-4. 

Symoneaux, R., Gahnarini, M. V. and Mehinagic, E. (2012) Comment analysis of consumer s likes and dislikes as an alternative tool to preference mapping. A case study on apples, Food Qual Prefer, 24, 59-66. DOI 10.1016/j.foodqual.2011.08.013. 

Varela, P., Beltran, J. and Fiszman, S. (2014) An alternative way to uncover drivers of coffee liking Preference mapping based on consumers preference ranking and open comments, Food Qual Prefer, 32B, 152-159. DOI 10.1016/j.foodqual.2013.03.004. 

Paulsen, M. T., Naes, T., Ueland, O., Rukke, E.-O. and Hersleth, M. (2013). Preference mapping of sahnon-sauce combinations The influence of temporal properties. Food Quality and Preference, n 2), 120-127. 

The concepts, as well as the corresponding step-by-step methodology for the analysis of the IPM data (called the Ideal Profile Analysis (IPA)), will be presented. Finally, the advantages/inconveniences of the IPM and its practical use compared to other methods (such as the Preference Mapping or JAR scale) will be discussed. 

The IPM is a useful tool for product optimization and product development and can be applied to a large variety of products (food, beverages, cosmetics, etc.). Compared to other optimization tools (such as external preference mapping, or LSA for example), it has the advantage that a large variety of information is gathered directly from the same consumers. This large variety of data allows the user to ... 

It is worth saying that external preference mapping techniques and tasks involving JAR questions are able to deal with multiple ideals, either by estimating different acceptance areas in the product space for a given consumer (PrefMap) or by providing guidance on improvement for each product separately. 

Worch, T. (2013) PrefMFA, a solution taking the best of both internal and external preference mapping techiuques, Food Quality and Pr erence, 30, 181-191. 

When consumer data are also available on the same products, we can build up an External Preference Mapping model to visualize how consumer preferences are linked to the sensory characteristics of the products (Schlich, 1995). This tool is especially powerful in orientating further product development towards consumer appreciation. Figure 15.3 shows an example of Preference Mapping, relating the sensory characteristics of eight European toast breads from different countries to the preference of Spanish consumers for the same toast breads. We can see that consumers globally prefer softer and moister products. The Spanish bread (actually the brand of the market leader) is close to this profile, so it is well in line with the taste expectations of the Spanish consumers. 

Figure 15.3 Preference Mapping of European toast breads for Spanish consumers. Zone A corresponds to higher consumer preferences. Zone B corresponds to lower consumer preferences.

Schlich, P. (1995) Preference Mapping Relating consumer preferences to sensory or instrumental measurements, in Etievant P and Schrier P, Bioflavour, INRA Dijon, 135-150. 

Danzart, M., Sieffermann, J.-M. and Delarne, J. (2004) New developments in preference mapping techniqnes Finding ont a consnmer optimal prodnct, its sensory profile and the key sensory attribntes. 6th Sensometric Meeting, Davis, CA USA. 

Although the focus of this chapter is on descriptive analysis, the consumer study will be presented because the results will emphasize the fact that even rapid descriptive methods can be used to understand consumers perceptions, through External Preference Mapping for instance. 

Our objective is to obtain a sensory description of the cars in a situation where roll and lateral support will be highly perceptible. As this description should be used for preference mapping, we initially aimed at conducting a conventional profile and we designed the first tests accordingly. 

On one hand, as we stiU have the objective of preference mapping, a minimum set of consensual sensations is required. On the other hand, we wanted to preserve the inter-individual differences of perception between the participants. Consequently, our methodological approach has been based on both individual and group sessions. The final list of sensations is composed of terms used by all participants, and terms used by just part of the group. 

To pursue our objective to understand the impact of roll and lateral support on the road-holding sensation, we used External Preference Mapping with quadratic model. The individual models of the 151 drivers with the consensus sensory map are quite good the average coefficient is 0.87 (Fig. 20.13). As anticipated with the raw results of the consumer study, the best area in terms of preference is near RC3-S3 but this is not the main point we would like to raise here. 

External Preference Mapping with quadratic model on 151 consumers... 

---