Feature engineering is required to obtain better results for time series forecasting, and decomposition is a crucial one. One decomposition approach often cannot be used for numerous forecasting tasks since the standard time series decomposition lacks flexibility and robustness. Traditional feature selection relies heavily on preexisting domain knowledge, has no generic methodology, and requires a lot of labor. However, most time series prediction models based on deep learning typically suffer from interpretability issues, so the “black box” results lead to a lack of confidence. Dealing with the above matters forms the motivation of the thesis. In the paper, we propose TSDFNet as a neural network with a self-decomposition mechanism and an attentive feature fusion mechanism. It abandons feature engineering as a preprocessing convention and creatively integrates it as an internal module with the deep model. The self-decomposition mechanism empowers TSDFNet with extensible and adaptive de-composition capabilities for any time series and is more effective in predicting problems than traditional decomposition methods. Users can choose their own basis functions to decompose the sequence into temporal and generalized spatial dimensions. The attentive feature fusion mechanism has the ability to capture the importance of external variables and the causality with target variables. It can automatically suppress the unimportant features while enhancing the effective ones so that users do not have to struggle with feature selection. The basic idea of TDFNNet is straightforward. It integrates feature decomposition, feature selection, and prediction into one model. Moreover, TSDFNet can quickly look into the “black box” of the deep neural network by feature visualization and analyzing the prediction results. We demonstrate performance improvements over existing widely accepted models on more than a dozen datasets, and three experiments showcase the interpretability of TSDFNet.

We present a deep learning-based thumbnail generation method called CropNet in this paper. Unlike previous deep learning-based methods, such as Fast-AT, which can utilize detectors introduced in object detection frameworks and generate thousands of proposals, our detector is straightforward and concise, thereby ensuring that the final cropping window is computed by its center and width, with the input aspect ratio. To achieve this goal, CropNet learns specific filters to estimate the center position and utilizes a cascade structure of filters and single neuron for width inference. In addition, CropNet optimizes the center and width jointly for optimal results. We collect a data set of more than 29,000 thumbnail annotations to train CropNet and perform cross-validation between existing data sets. Experiments show that CropNet outperforms existing techniques. Our result is achieved at a test-time speed of 17 ms per image, which is six times faster than the fastest method at present.