Source Themes

3DACN, 3D Augmented Convolutional Network for Time Series Data
In this paper, we propose 3D Augmented Convolutional Network(3DACN) for hybird time series data by setting three dimensionsy to make convolutions along time dimension and use augmented algorithm and EM algorithm to solve imbalanced data problem and extract the value concatenation containing time series information respectivly. We verified the 3DACN on two Databases and got better performance of F1 score and AUC than other algorithms. According to the result of F1 score and AUC, we can prove our 3DACN having a strong ability to deal with hybird time series data including imbalance data.
Songwen Pei, Tianma Shen, Chunhua Gu, Bingxue Zhang, Zhong Ning, Naixue Xiong
Cite
3DACN, 3D Augmented Convolutional Network for Time Series Data