Predicting Grain Losses and Waste Rate Along the Entire Chain: A Multitask Multigated Recurrent Unit Autoencoder Based Method

被引:21
作者
Cao, Jie [1 ]
Wang, Youquan [1 ]
He, Jing [1 ]
Liang, Weichao [2 ]
Tao, Haicheng [1 ]
Zhu, Guixiang [1 ]
机构
[1] Nanjing Univ Finance & Econ, Jiangsu Prov Key Lab E Business, Coll Informat Engn, Nanjing 210046, Peoples R China
[2] Nanjing Univ Sci & Technol, Coll Comp Sci & Engn, Nanjing 210094, Peoples R China
基金
中国国家自然科学基金;
关键词
Task analysis; Predictive models; Informatics; Interviews; Economics; Loss measurement; Deep learning; grain losses and waste rate (LWR) prediction; multitask prediction; recurrent skip connection network (RSCN); FOOD WASTE;
D O I
10.1109/TII.2020.3030709
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Predicting grain losses and waste rate (LWR) is critical for agricultural planning and grain policy development. Capturing the stage interaction and generating robust features are the main challenges in grain LWR prediction. In this article, we propose MTGA, a Multitask Gated recurrent unit (GRU) Autoencoder, approach to 1) obtain the robust feature representation for the prediction task and 2) explore the time-ordered interactions among different stages of the grain chain. Specifically, we design multiple GRU encoder-decoder pairs to co-reconstruct the stage features in a common space for robust feature learning. Then, an attention mechanism is proposed better to fuse the reconstructed features from the GRU encoder-decoder pairs. Furthermore, we utilize the multitask for reconstructed loss and grain LWR prediction. We introduce the reconstructed loss task as an auxiliary task to help us to represent the robust features. Besides, we introduce the LWR prediction as main task to learn the parameters for prediction task. We collected the data with questionnaires, interviews, or data from grain management institutes for experiments. The evaluation results show that grain LWR prediction by our approach achieves the best results compared to several state-of-the-art prediction models. Moreover, our method gains overall performance decline of 12.5-18.3% on mean absolute error and root mean square error metrics.
引用
收藏
页码:4390 / 4400
页数:11
相关论文
共 36 条
[31]  
Wang Y., 2016, EMNLP, P938
[32]   Real-Time Identification of Power Fluctuations Based on LSTM Recurrent Neural Network: A Case Study on Singapore Power System [J].
Wen, Shuli ;
Wang, Yu ;
Tang, Yi ;
Xu, Yan ;
Li, Pengfei ;
Zhao, Tianyang .
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2019, 15 (09) :5266-5275
[33]   hPSD: A Hybrid PU-Learning-Based Spammer Detection Model for Product Reviews [J].
Wu, Zhiang ;
Cao, Jie ;
Wang, Yaqiong ;
Wang, Youquan ;
Zhang, Lu ;
Wu, Junjie .
IEEE TRANSACTIONS ON CYBERNETICS, 2020, 50 (04) :1595-1606
[34]   Robust Task Grouping with Representative Tasks for Clustered Multi-Task Learning [J].
Yao, Yaqiang ;
Cao, Jie ;
Chen, Huanhuan .
KDD'19: PROCEEDINGS OF THE 25TH ACM SIGKDD INTERNATIONAL CONFERENCCE ON KNOWLEDGE DISCOVERY AND DATA MINING, 2019, :1408-1417
[35]   Multitask Autoencoder Model for Recovering Human Poses [J].
Yu, Jun ;
Hong, Chaoqun ;
Rui, Yong ;
Tao, Dacheng .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2018, 65 (06) :5060-5068
[36]  
Zhang Y., 2017, CoRR, abs/1707.08114