Sensory data analysis

Procrustes analysis has been generalized in two ways. One extension is that more than two data sets may be considered. In that case the algorithm is iterative. One then must rotate, in turn, each data set to the average of the other data sets. The cycle must be repeated until the fit no longer improves. Procrustes analysis of many data sets has been applied mostly in the field of sensory data analysis [4]. Another extension is the application of individual scaling to the various data sets in order to improve the match. Mathematically, it amounts to multiplying all entries in a data set by the same scalar. Geometrically, it amounts to an expansion (or... 

The analysis was performed using the statistical software R, the FactoMineR package, an R package dedicated to exploratory multivariate analysis, and the SensoMineR package, an R package dedicated to sensory data analysis. 

Le, S. and Husson, F. (2008) SensoMineR a package for sensory data analysis, Journal of Sensory Studies, 23, 14—25. 

Let us try to relate the (standardized) sensory data in Table 35.1 to the explanatory variables in Table 35.3. Essentially, this is an analysis-of-variance problem. We try to explain the effects of two qualitative factors, viz. Country and Ripeness, on the sensory responses. Each factor has three levels Country = Greece, Italy,... 

As an example we try to model the relation between the sensory data of Table 35.1 and the instmmental measurements of Table 35.4. The PLS analysis results are shown in Table 35.8. The first PLS dimension loads about equally high on... 

The determination and analysis of sensory properties plays an important role in the development of new consumer products. Particularly in the food industry sensory analysis has become an indispensable tool in research, development, marketing and quality control. The discipline of sensory analysis covers a wide spectrum of subjects physiology of sensory perception, psychology of human behaviour, flavour chemistry, physics of emulsion break-up and flavour release, testing methodology, consumer research, statistical data analysis. Not all of these aspects are of direct interest for the chemometrician. In this chapter we will cover a few topics in the analysis of sensory data. General introductory books are e.g. Refs. [1-3]. 

Beilken et al. [ 12] have applied a number of instrumental measuring methods to assess the mechanical strength of 12 different meat patties. In all, 20 different physical/chemical properties were measured. The products were tasted twice by 12 panellists divided over 4 sessions in which 6 products were evaluated for 9 textural attributes (rubberiness, chewiness, juiciness, etc.). Beilken etal. [12] subjected the two sets of data, viz. the instrumental data and the sensory data, to separate principal component analyses. The relation between the two data sets, mechanical measurements versus sensory attributes, was studied by their intercorrelations. Although useful information can be derived from such bivariate indicators, a truly multivariate regression analysis may give a simpler overall picture of the relation. 

P. Lea, T. Naes and M. R0dbotton, Analysis of Variance for Sensory Data. Wiley, London, 1997 D. H. Lyon, M. A. Francombe, T. A. Hasdell and K. Lawson, Guidelines for Sensory Analysis in Product Development and Quality Control. Chapman and Hall, London, 1990. 

Multivariate Analysis of Data in Sensory Science, edited by T. Naes and E. Risvik Data Analysis for Hyphenated Techniques, by E.J. Karjalainen and U.P. Karjalainen... 

Volume 16 Multivariate Analysis of Data in Sensory Science, edited by T. Naes and E. Risvik Volume 17 Data Analysis for Hyphenated Techniques, by E.J. Karjalainen and U.P. Karjalainen Volume 18 Signal Treatment and Signal Analysis in NMR, edited by D.N. Rutledge Volume 19 Robustness of Analytical Chemical Methods and Pharmaceutical Technological Products, edited by M.M.W.B. Hendriks, J.H. de Boer and A.K. Smilde... 

Thybo, A. K., Martens, M. (1998). Development of a sensory texture profile of cooked potatoes by multivariate data analysis. Journal of Texture Studies, 29,453 68. 

Since the sensory data collected involved degree of sample difference from a reference, it was felt that the analytical data should be analyzed in a similar manner. In cases where some peaks making up a multicomponent mixture are known to be specific to that mixture, this is a relatively simple matter. In such cases, the peak areas of the known components can be compared to a reference and average percent difference calculated. However, if it is not possible to pick out peaks that are clearly specific to a single multicomponent mixture, a more sophisticated technique such as factor analysis is required. There are circumstances where all peaks are common to each multicomponent mixture, i.e. qualitatively similar but quantitatively different. Also there are cases where peaks are found only in one of the multicomponent mixtures, but it is not clear to which mixture they belong. In these cases factor analysis is required to extract patterns that are characteristic of the specific multicomponent mixtures. Analytical concentrations of each of the multicomponent mixtures are then calculated as a set of factor scores where each score is directly proportional to the actual concentration of each multicomponent mixture. 

Correlation of Analytical/Sensory Results. Sensory data was correlated with headspace data of tobacco volatiles by factor analysis (BMDP4M) and canonical correlation BMDP6M. Analytical data included factor scores and discriminant analyses scores sensory data included scores from the two MDS dimensions. Sorted rotated factor loadings of combined sensory/analytical data using factor analysis are shown in Table II. Factor one contained those variables from the analytical and sensory data which related to differences between bright (A), burley (B), and oriental (C) (Figure 10). These included dimension 1 in the... 

Both factor analysis and canonical correlation techniques were successful in demonstrating that differences between tobacco type and casing could be detected from both analytical and sensory data, and that those differences found analytically were highly correlated to sensory differences. From this type of data correlation, components can be pinpointed which may be responsible for sensory differences between tobacco types. 

The two research investigations reported here - the sensory quality control specification model and the application of sensory and analytical data for defining differences in tobacco aroma - both demonstrate the usefulness of multivariate analysis techniques for analyzing analytical and sensory data as well as correlating these data. Although these tasks do not compare in complexity to that of the prediction of sensory response to analytical data collected on cigarette smoke, our research to date has revealed no element which indicates that this is an impossible task. In fact, the results of these and similar... 

Dupuy and coworkers have reported a direct gas chromatographic procedure for the examination of volatiles in vegetable oils (11). peanuts and peanut butters (12, 13), and rice and com products (14). When the procedure was appTTed to the analysis of flavor-scored samples, the instrumental data correlated well with sensory data (15, 16, 17), showing that food flavor can be measured by instrvmental means. Our present report provides additional evidence that the direct gas chromatographic method, when coupled with mass spectrometry for the identification of the compounds, can supply valid information about the flavor quality of certain food products. Such information can then be used to understand the mechanisms that affect flavor quality. Experimental Procedures... 

Almost everyone is now utilizing the computer for statistical analysis of sensory data. Some laboratories also are using computers to gather the data as well (18). A computerized sensory system would benefit most laboratories by freeing workers from laborious data entry and analysis. Also, it would allow for a more thorough analysis of the data. It should not replace inspection of the raw data by the sensory scientist, but allow this to occur more easily. 

Glucose was the only major sugar and IMP and GMP were the only major nucleotides found. A sensory evaluation of the different processed products Indicated a preference for the drum dried product over the freeze or spray dried product. This preference could not be explained from sugar or nucleotide values and the amino acid data was Inconclusive. Since the authors have amassed such a large data pool for both volatile and nonvolatile compounds 1t Is unfortunate that some form of data analysis such as multlvarlent statistical analysis was not applied so as to determine which compounds were primarily responsible for the perceived flavor preference. 

In a recent study, using multivariate statistical analysis of quantitative sensory descriptive analysis and precise chemical compositional data, Smyth et al. (2005) found that the importance of individual yeast esters to the aroma profile of wine can vary with the type of wine. In the case of unwooded Chardonnay wines, for... 

The published literature on the effects of microbial activities on wine chemical composition is now considerable. Understanding the significance of wine chemistry is, however, heavily dependent on complex analytical strategies which combine extensive chemical characterization and sensory descriptive analysis. However, sensory analysis is extremely resource-intense, requiring many hours of panelists time. This prevents widespread application of these powerful analytical tools. Advanced statistical techniques have been developed that are closing the gap between chemical and sensory techniques. Such techniques allow the development of models, which should ultimately provide a sensory description based on chemical data. For example, Smyth et al. (2005) have developed reasonable models which can reveal the most likely compounds that relate to particular attributes that characterise the overall sensory profile of a wine. For wines such as Riesling and Chardonnay, the importance of several yeast volatile compounds has been indicated. Such information will allow yeast studies to target key compounds better rather than just those that are convenient to measure. 

Acid hydrolysates were added to a low aroma intensity white wine (ie the base wine), and the aroma properties of these samples were assessed by sensory descriptive analysis. In addition, the glycoside isolates from the Australian vineyards were subjected to glycoside hydrolase enzyme treatment, and duo-trio difference tests were performed on these hydrolysates added to a base wine. The volatile composition of each of the hydrolysates was investigated by GC/MS, and relationships between the two sets of data were determined. Finally, the glycoside concentration of each of the juices and skin extracts was determined by the glycosyl-glucose assay. 

Sensory analysis. Significant differences in intensity were found for all seven aroma terms by analysis of variance (data not shown). Because of a highly significant judge-by-wine interaction, the berry term was excluded from further data analysis. 

Figure 1. Sensory descriptive analysis data of Napa Cabernet Sauvignon samples and the base wine. Mean ratings of 14 judges x 2 replicates and least significant differences (LSD, p<0.05) are shown. For sample codes, see Table II.

---