Novel computational tools in bakery process data analysis

A comparative study

Juho Rousu (Corresponding Author), Laura Flander, Marjaana Suutarinen, Karin Autio, Petri Kontkanen, Ari Rantanen

Research output: Contribution to journalArticleScientificpeer-review

22 Citations (Scopus)

Abstract

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.
Original languageEnglish
Pages (from-to)45-56
Number of pages12
JournalJournal of Food Engineering
Volume57
Issue number1
DOIs
Publication statusPublished - 2003
MoE publication typeA1 Journal article-refereed

Fingerprint

pastries
data analysis
rye
breads
least squares
Bread
sourdough
Least-Squares Analysis
prediction
artificial intelligence
pruning
product quality
neural networks
statistical analysis
learning
methodology
Decision Trees
Secale
Machine Learning
support vector machines

Keywords

  • Data analysis
  • Predictive modeling
  • Bakery processes
  • Product quality

Cite this

Rousu, J., Flander, L., Suutarinen, M., Autio, K., Kontkanen, P., & Rantanen, A. (2003). Novel computational tools in bakery process data analysis: A comparative study. Journal of Food Engineering, 57(1), 45-56. https://doi.org/10.1016/S0260-8774(02)00221-2
Rousu, Juho ; Flander, Laura ; Suutarinen, Marjaana ; Autio, Karin ; Kontkanen, Petri ; Rantanen, Ari. / Novel computational tools in bakery process data analysis : A comparative study. In: Journal of Food Engineering. 2003 ; Vol. 57, No. 1. pp. 45-56.
@article{9ff2174cdcd6484daf170385d3a3baef,
title = "Novel computational tools in bakery process data analysis: A comparative study",
abstract = "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.",
keywords = "Data analysis, Predictive modeling, Bakery processes, Product quality",
author = "Juho Rousu and Laura Flander and Marjaana Suutarinen and Karin Autio and Petri Kontkanen and Ari Rantanen",
year = "2003",
doi = "10.1016/S0260-8774(02)00221-2",
language = "English",
volume = "57",
pages = "45--56",
journal = "Journal of Food Engineering",
issn = "0260-8774",
publisher = "Elsevier",
number = "1",

}

Rousu, J, Flander, L, Suutarinen, M, Autio, K, Kontkanen, P & Rantanen, A 2003, 'Novel computational tools in bakery process data analysis: A comparative study', Journal of Food Engineering, vol. 57, no. 1, pp. 45-56. https://doi.org/10.1016/S0260-8774(02)00221-2

Novel computational tools in bakery process data analysis : A comparative study. / Rousu, Juho (Corresponding Author); Flander, Laura; Suutarinen, Marjaana; Autio, Karin; Kontkanen, Petri; Rantanen, Ari.

In: Journal of Food Engineering, Vol. 57, No. 1, 2003, p. 45-56.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Novel computational tools in bakery process data analysis

T2 - A comparative study

AU - Rousu, Juho

AU - Flander, Laura

AU - Suutarinen, Marjaana

AU - Autio, Karin

AU - Kontkanen, Petri

AU - Rantanen, Ari

PY - 2003

Y1 - 2003

N2 - 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.

AB - 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.

KW - Data analysis

KW - Predictive modeling

KW - Bakery processes

KW - Product quality

U2 - 10.1016/S0260-8774(02)00221-2

DO - 10.1016/S0260-8774(02)00221-2

M3 - Article

VL - 57

SP - 45

EP - 56

JO - Journal of Food Engineering

JF - Journal of Food Engineering

SN - 0260-8774

IS - 1

ER -