A Multitask Learning Framework for Multi-Property Detection of Wine

The electronic nose (E-nose) is a bionic olfactory system and a powerful tool in many fields. Sample classification and parameter prediction are the core functions of the E-nose. We present two algorithms for simultaneous recognition of four properties (wine region, grape variety, vintage, and fermentation processes) based on a back-propagation neural network (BPNN) and convolutional neural network (CNN), respectively, where the tasks (i.e., identification of the four properties) share underlying features. These algorithms exploited synergy among tasks to enhance their individual performance. Experimental results show that the model based on BPNN achieved the best performance with accuracies of 94.5%, 83.7%, 75.1%, and 76.9% in identifying wine region, grape, vintage, and fermentation processes, respectively. Furthermore, the results reveal that the models can capture global and local information and perform better than single-task models.

[1]  V. Rico-Ramírez,et al.  Gas chromatography/mass spectrometry for the determination of nitrosamines in red wine. , 2016, Food chemistry.

[2]  Qiang Li,et al.  Application of Random Forest Classifier by Means of a QCM-Based E-Nose in the Identification of Chinese Liquor Flavors , 2017, IEEE Sensors Journal.

[3]  Yu Gu,et al.  Bionic Electronic Nose Based on MOS Sensors Array and Machine Learning Algorithms Used for Wine Properties Detection , 2018, Sensors.

[4]  Paul Geladi,et al.  Principal Component Analysis , 1987, Comprehensive Chemometrics.

[5]  Rich Caruana,et al.  Multitask Learning , 1997, Machine-mediated learning.

[6]  Jiliang Zhang,et al.  On Modified Multi-Output Chebyshev-Polynomial Feed-Forward Neural Network for Pattern Classification of Wine Regions , 2019, IEEE Access.

[7]  Yu Gu,et al.  Classification of Multiple Chinese Liquors by Means of a QCM-based E-Nose and MDS-SVM Classifier , 2017, Sensors.

[8]  Calum Morrison,et al.  Direct solid phase microextraction combined with gas chromatography - Mass spectrometry for the determination of biogenic amines in wine. , 2018, Talanta.

[9]  Gerald Penn,et al.  Convolutional Neural Networks for Speech Recognition , 2014, IEEE/ACM Transactions on Audio, Speech, and Language Processing.

[10]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[11]  Zuobing Xiao,et al.  Application of Gas Chromatography-Based Electronic Nose for Classification of Chinese Rice Wine by Wine Age , 2014, Food Analytical Methods.

[12]  Xuemei Zhou,et al.  Network traffic prediction based on BPNN optimized by self-adaptive immune genetic algorithm , 2013, Proceedings 2013 International Conference on Mechatronic Sciences, Electric Engineering and Computer (MEC).

[13]  Xuanjing Huang,et al.  Recurrent Neural Network for Text Classification with Multi-Task Learning , 2016, IJCAI.

[14]  Rama Chellappa,et al.  HyperFace: A Deep Multi-Task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition , 2019, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Cheng Han,et al.  Hybrid approach for human posture recognition using anthropometry and BP neural network based on Kinect V2 , 2019, EURASIP J. Image Video Process..

[16]  Dean Zhao,et al.  An optimized classification algorithm by BP neural network based on PLS and HCA , 2014, Applied Intelligence.

[17]  N. E. Bari,et al.  E-Nose and e-Tongue combination for improved recognition of fruit juice samples. , 2014, Food chemistry.

[18]  Fengye Hu,et al.  A human body posture recognition algorithm based on BP neural network for wireless body area networks , 2016, China Communications.

[19]  Xiaojun Chang,et al.  Semisupervised Feature Analysis by Mining Correlations Among Multiple Tasks , 2014, IEEE Transactions on Neural Networks and Learning Systems.

[20]  Daniel Cozzolino,et al.  Classification of Tempranillo wines according to geographic origin: combination of mass spectrometry based electronic nose and chemometrics. , 2010, Analytica chimica acta.

[21]  Antonello Rizzi,et al.  Short-Term Electric Load Forecasting Using Echo State Networks and PCA Decomposition , 2015, IEEE Access.

[22]  F. Zhang,et al.  Direct determination of multi-pesticides in wine by ambient mass spectrometry , 2017 .

[23]  Yang Wei,et al.  Application of Electronic Nose for Detection of Wine-Aging Methods , 2014 .

[24]  Guo-Wei Wei,et al.  Quantitative Toxicity Prediction Using Topology Based Multitask Deep Neural Networks , 2017, J. Chem. Inf. Model..

[25]  J. Welke,et al.  Sensory, olfactometry and comprehensive two-dimensional gas chromatography analyses as appropriate tools to characterize the effects of vine management on wine aroma. , 2018, Food chemistry.

[26]  Yennifer Yuliana Rios Diaz,et al.  Electronic tongue for wine discrimination, using PCA and ANN , 2014 .

[27]  José Ignacio Suárez,et al.  Automatic Sensor System for the Continuous Analysis of the Evolution of Wine , 2015, American Journal of Enology and Viticulture.

[28]  Dong Yu,et al.  Exploring convolutional neural network structures and optimization techniques for speech recognition , 2013, INTERSPEECH.

[29]  Cristina Medina-Plaza,et al.  Electronic Noses and Tongues in Wine Industry , 2016, Front. Bioeng. Biotechnol..

[30]  Wei Li,et al.  Electronic nose with a new feature reduction method and a multi-linear classifier for Chinese liquor classification. , 2014, The Review of scientific instruments.

[31]  Xin Pan,et al.  A hybrid MLP-CNN classifier for very fine resolution remotely sensed image classification , 2017, ISPRS Journal of Photogrammetry and Remote Sensing.

[32]  Yi Wan,et al.  Chinese Wine Classification Using BPNN through Combination of Micrographs' Shape and Structure Features , 2009, 2009 Fifth International Conference on Natural Computation.