Novel computational tools in bakery process data analysis: a comparative study

We studied the potential of various machine learning and statistical methods in the prediction of product quality in industrial bakery processes. The methods included classification and regression tree, decision list, neural network, support vector machine and Bayesian learning algorithms as well as statistical multivariate methods. Our data originated from two industrial bakery processes: a sourdough rye bread and a Danish pastry process. In our studies, the Naive Bayesian algorithm turned out to be the best classifier building algorithm while the partial least squares (PLS) method was the best regression method. The prediction accuracy of these models improved significantly by pruning the original set of variables. In this study, two response variables could be predicted on a level that justifies further study: rye bread pH could be predicted with high accuracy with Naive Bayesian Classifier, and Danish pastry height could be predicted with a moderately high correlation with PLS.

[1]  Søren Balling Engelsen,et al.  NMR-baking and multivariate prediction of instrumental Texture parameters in bread , 2001 .

[2]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[3]  Ellen Mosleth Færgestad,et al.  Effects of flour composition, physical dough properties and baking process on hearth loaf properties studied by multivariate statistical methods. , 2000 .

[4]  Thomas M. Loughin,et al.  Use of regression and discriminant analyses to develop a quality classification system for hard red winter wheat , 1999 .

[5]  S. Wold,et al.  The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized Inverses , 1984 .

[6]  Bert L. D'Appolonia,et al.  The Farinograph Handbook , 1984 .

[7]  Per Åman,et al.  Predictive Modelling of the Bread-making Performance and Dough Properties of Wheat , 1994 .

[8]  Bernhard Schölkopf,et al.  Learning with kernels , 2001 .

[9]  Susan Linko,et al.  Expert systems—what can they do for the food industry? , 1998 .

[10]  J. Ross Quinlan,et al.  Combining Instance-Based and Model-Based Learning , 1993, ICML.

[11]  Gilles Trystram,et al.  Sensor fusion for real time quality evaluation of biscuit during baking. Comparison between Bayesian and fuzzy approaches , 1996 .

[12]  F. Schierbaum,et al.  W. C. SHUEY und K. H. TIPPLES: The Amylograph Handbook. 37 Seiten, zahlr. Abb. und Tab. The American Association of Cereal Chemists, St. Paul, Minnesota, 1980. , 1982 .

[13]  L. Bohlin,et al.  Multivariate analysis as a tool to predict bread volume from mixogram parameters , 1996 .

[14]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[15]  R. Rivest.  Learning Decision Lists , 1987, Machine Learning.

[16]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[17]  John R. Mitchell,et al.  A model for the oven rise of dough during baking , 1999 .

[18]  Peter Clark,et al.  The CN2 Induction Algorithm , 1989, Machine Learning.

[19]  Karin Autio,et al.  Influence of starch granule size distribution on bread characteristics , 1998 .

[20]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[21]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[22]  O. M. Lukow,et al.  Neural Networks vs Principal Component Regression for Prediction of Wheat Flour Loaf Volume in Baking Tests , 1995 .

[23]  Ronald L. Rivest,et al.  Training a 3-node neural network is NP-complete , 1988, COLT '88.

[24]  Alexander J. Smola,et al.  Support Vector Machine Reference Manual , 1998 .

[25]  C. K. Spillman,et al.  Neural network modeling of physical properties of ground wheat. , 1998 .