Name
Title | ||
|---|---|---|
Time Series Pattern Classification Based on Deep Learning Network - A Study on the Roles of Time Domain Representation. |
Abstract | ||
|---|---|---|
Recently, deep learning network is the hottest machine learning technology in the pattern recognition community. In past years, preprocessing of the raw data is always needed to reduce the dimensions of the input data so that the learning process can be completed within a reasonable time. However, the great improvement of the computational power allows us to adopt the raw data to serve as the input directly for the deep learning algorithm. Besides the ability for dimensionality reduction, the behaviors, especially the benefits, of representing the time series patterns in time domain for classification are studied in this research. It is expected that time domain time series pattern representation has its own strengths for the time series data mining problems. |
Year | DOI | Venue |
|---|---|---|
2018 | 10.1109/FSKD.2018.8687275 | ICNC-FSKD |
Field | DocType | Citations |
Time domain,Time series data mining,Dimensionality reduction,Computer science,Raw data,Preprocessor,Artificial intelligence,Deep learning,Machine learning | Conference | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Tak-Chung Fu | 1 | 0 | 0.34 |
| Ying Kit Hung | 2 | 0 | 0.34 |
| Fu-lai Chung | 3 | 244 | 34.50 |