An ensemble of AHP-EW and AE-RNN for food safety risk early warning
Jie Zhong,
Lei Sun,
Enguang Zuo,
Cheng Chen,
Chen Chen,
Huiti Jiang,
Hua Li and
Xiaoyi Lv
PLOS ONE, 2023, vol. 18, issue 4, 1-26
Abstract:
Food safety problems are becoming increasingly severe in modern society, and establishing an accurate food safety risk warning and analysis model is of positive significance in avoiding food safety accidents. We propose an algorithmic framework that integrates the analytic hierarchy process based on the entropy weight (AHP-EW) and the autoencoder-recurrent neural network (AE-RNN). Specifically, the AHP-EW method is first used to obtain the weight percentages of each detection index. The comprehensive risk value of the product samples is obtained by weighted summation with the detection data, which is used as the expected output of the AE-RNN network. The AE-RNN network is constructed to predict the comprehensive risk value of unknown products. The detailed risk analysis and control measures are taken based on the risk value. We applied this method to the detection data of a dairy product brand in China for example validation. Compared with the performance of 3 models of the back propagation algorithm (BP), the long short-term memory network (LSTM), and the LSTM based on the attention mechanism (LSTM-Attention), the AE-RNN model has a shorter convergence time, predicts data more accurately. The root mean square error (RMSE) of experimental data is only 0.0018, proving that the model is feasible in practice and helps improve the food safety supervision system in China to avoid food safety incidents.
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations:
Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0284144 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 84144&type=printable (application/pdf)
Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.
Export reference: BibTeX
RIS (EndNote, ProCite, RefMan)
HTML/Text
Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0284144
DOI: 10.1371/journal.pone.0284144
Access Statistics for this article
More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().