Deep autonomous learning machine for IoT streaming analytics
With the technological advancement in Internet of Things (IoT), it has been employed in multiple areas such as healthcare, manufacturing, home automations. The wide applications of IoT accelerate the rate of data generation, resulting in an explosion of data. Furthermore, IoT devices generate and tr...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Final Year Project |
| 語言: | English |
| 出版: |
Nanyang Technological University
2020
|
| 主題: | |
| 在線閱讀: | https://hdl.handle.net/10356/138224 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Nanyang Technological University |
| 語言: | English |
| 總結: | With the technological advancement in Internet of Things (IoT), it has been employed in multiple areas such as healthcare, manufacturing, home automations. The wide applications of IoT accelerate the rate of data generation, resulting in an explosion of data. Furthermore, IoT devices generate and transmit data in streams which lead to increase interest in real-time data stream classification. Consequently, traditional algorithms are inefficient to cope with the large volumes of data stream. In various IoT applications, the use of recurrent neural network (RNN) is desirable due to the sequential nature of the data stream. This allows an RNN-based classifier to handle sequential data and to retain temporal information. However, with the large volume of data stream, traditional RNN-based classifiers are offline in nature and impractical in the streaming context. ADL and NADINE are therefore, introduced to address data stream problems in the continual fashion. The ADL and NADINE are compared with traditional RNN-based classifiers to demonstrates their performance in data stream classification. |
|---|