WJWentao Jiang

Wentao Jiang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Wentao Jiang has a total of 14 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Two forecasting model selection methods based on time series image feature augmentation

    by Wentao Jiang, Quan Wang, Hongbo Li

    Forecasting and early warning of agricultural product prices is a crucial task in stream data event analysis and agricultural data mining. Existing methods for forecasting agricultural product prices suffer from inefficient feature engineering and challenges in handling imbalanced sample data. To address these issues, we propose a novel predictive model selection approach based on time series image encoding. Specifically, we utilize Gramian Angular Fields (GAF), Markov Transition Fields (MTF), and Recurrence Plots (RP) to transform time series data into image representations. We then introduce an Information Fusion Feature Augmentation (IFFA) method to effectively combine these time series images, ensuring that all relevant event information is preserved. The combined time series images (TSCI) are subsequently fed into a Convolutional Neural Network (CNN) classifier for model selection. Furthermore, to accommodate the unique characteristics of the data, we incorporate Transfer Learning (TL) and S-Folder Cross Validation (S-FCV) to optimize the model selection process, thereby mitigating overfitting due to limited or imbalanced data. Experimental results demonstrate that the proposed IFFA-TSCI-CNN-SFCV method outperforms existing approaches in terms of both efficiency and accuracy.


  • Open Access

    Time series image coding classification theory based on Lagrange multiplier method

    by Wentao Jiang, Ming Zhao, Hongbo Li

    Time series classification is a crucial area of research within data analysis. Recent advancements in convolutional neural networks (CNNs) for image recognition and classification have inspired innovative approaches in this field. Researchers have begun transforming one-dimensional time series (TS) into visual representations, harnessing the capabilities of image classification paradigms to improve TS classification accuracy. Despite extensive research on the classification of time series images (TSI), the underlying classification principles have not been fully established. To address this gap, our paper integrates time series image encoding with algebraic techniques, utilizing Gramian Angular Summation Fields (GASF) and Gramian Angle Difference Fields (GADF) as effective encoding methods. We provide a theoretical justification for the advantages of CNNs in TSI classification by employing the Lagrange multiplier method and the Karush-Kuhn-Tucker (KKT) conditions. This rigorous proof framework enhances the theoretical foundation for TSI classification based on image encoding. To validate our theoretical assertions, we conduct experiments using the benchmark UCR dataset, with empirical findings largely corroborating our theoretical proofs. In summary, our work not only advances the theoretical understanding of TSI classification but also demonstrates its practical effectiveness.


  • Open Access

    Time series to imaging based deep learning model for detecting abnormal fluctuation in agriculture product price

    by Jiang Wentao, Dabin Zhang, Liwen Ling, Guotao Cai, Liling Zeng

    In the analysis of agricultural product price time series, the detection of abnormal fluctuation is the primary task. Accurately judging the abnormal fluctuation of agricultural product prices will give the policy support to the government and also assist farmers increase production and income. A deep convolutional neural network model based on time series image(TSI) is introduced to identify the abnormal fluctuation of agricultural prices under the improved standard deviation-Slope judgment. Markov Transfer Field(MTF) method is used to transform the pre-processed sparse one-dimensional time series of agricultural prices into two-dimensional dense images, and a deep convolutional neural network(CNN) model is used for automatic feature extraction and classification of time series images containing abnormal fluctuations. The empirical evaluation of China's corn and wheat price datasets are performed in our paper, and compared with other abnormal fluctuation judgment methods, the accuracy of the proposed algorithm is about 20% higher on average, which confirms the applicability of the standard deviation-slope time series Image-Resnet-34 (SDS-TSI-Resnet34) model in practical scenarios. Finally, some feasible suggestions for the efficient development of agricultural economy are proposed based on the abnormal fluctuation judgment method proposed in this paper.

  • Open Access

    A Time Series Forecasting Model Selection Framework Using CNN and Data Augmentation for Small Sample Data

    by Wentao Jiang, Liwen Ling, Dabin Zhang, Ruibin Lin, Liling Zeng

    The key to the accuracy of time series forecasting is to find the most appropriate forecasting method. Therefore, the forecasting model selection of time series has become a new research hotspot in the data analysis field. However, most of the existing forecasting model selection methods reduce the forecasting efficiency for relying on subjective manual selection of features. In this paper, an automatic time series feature extraction framework is proposed for forecasting model selection based on the idea of meta learning. Inspired by computer vision, we transform one -dimensional time series into two-dimensional images, and use convolution neural network (CNN) to train and classify time series images (model selection). Moreover, in order to deal with the over fitting problem caused by small sample datasets, the sliding window data augmentation method is used to improve the accuracy of small datasets model selection. A large-scale empirical study on M3 datasets shows that the proposed framework has better model selection accuracy and smaller forecasting error(MAPE) than Support vector machine(SVM) and traditional time series image algorithms. Moreover, the classification rate(model selection accuracy) of the proposed algorithm are increased by 6.5% and 4.4% compare with the traditional time series image method and Support vector machine respectively in average