Applying artificial neural networks as a test to detect milk fraud by whey addition
DOI:
https://doi.org/10.14295/2238-6416.v69i6.353Keywords:
classification, physicochemical analysis, RBFAbstract
This study aimed to employ Artificial Neural Networks to classify milk samples from routine analysis of a dairy company in order to identify adulteration by whey addition. The samples were prepared by mixing the milk with different whey concentrations (0, 1, 5, 10, and 20%), which were then analyzed for temperature, fat content, solids-non-fat, bulk density, protein, lactose, minerals, freezing point, conductivity, and pH, for a total of 167 assays. Out of these, 101 were used to train the network, 33 for validation, and 33 to test the artificial neural network. The best classification was obtained using a radial basis function neural network. k-means algorithm was used to obtain the network center, k-nearest was used to define the receptive fields, and the pseudo-inverse method was used to define the weights of the output layer. The best result was found with a network with 10 neurons in the input layer, 40 neurons in the hidden layer, and two neurons in the output layer, achieving over 95% accuracy in classification. The classification methodology using artificial neural networks has strong potential to be applied in interpreting data from routine analysis in dairy companies in order to classify milk adulterated with whey and, later, confirm the result using official methodologies.Downloads
Published
Issue
Section
License
Term of Responsibility and Transference of Author Rights
The undersigned authors of the article entitled "INSERT TITLE" declare to have read and approved all the manuscript. The authors agree to submit it to the Journal of Candido Tostes Dairy Institute for evaluation and possible publication as true, authentic, and original results. This statement implies that the manuscript, regardless of language, was not submitted to other journals with the same purpose. The authors acknowledge that will be required to provide retractions or corrections of errors related to the article, if necessary. The authors are aware of the guidelines and editorial policy of the Journal and assign copyright to the Journal of Candido Tostes Dairy Institute with the exclusive right to print, publish and sell the product worldwide, in all languages and media. Therefore, it is forbidden to authors reproduce wholly or partially the submitted manuscript in any other part or media, print or electronic. Thus, the authors declare that accepts to transfer the rights of graphic reproduction to the Journal of Candido Tostes Dairy Institute in the case of the article with the above title (or title that later come to be adopted, to meet the suggestions of the editors and reviewers) will be published in the Journal of Candido Tostes Dairy Institute. In addition, the authors agree to appoint _(name)___________ as the corresponding author.
This term should be signed by all authors in the order of authorship, stating:
Date;
Authors name without abbreviation;
Document number;
Title/position of each author;
e-mail; and
the area for evaluation of the paper.
