| Autor: |
Xi Zhang, Jiyue Wang, Ying Huang, Feiyue Zhu |
| Zdroj: |
Computer Science & Information Systems; Sep2024, Vol. 21 Issue 4, p1629-1649, 21p |
| Abstrakt: |
The traditional industrial big data fusion algorithm has low efficiency and difficulty in processing high-dimensional data, this paper proposes a Q-learningbased cascade classifier model for industrial big data fusion. By combining cascade classifier and softmax classifier, feature extraction and data attribute classification of source industrial big data are completed in this cluster. In order to improve the classification rate, an improved Q-learning algorithm is proposed, which makes the improved algorithm randomly select actions in the early stage, and dynamically change in the late stage in the random selection of actions and actions with the highest reward value. It effectively improves the defects of traditional Q-learning algorithm that it is easy to fall into the local optimal and has slow convergence speed. The experimental results show that compared with other advanced fusion algorithms, the proposed method can greatly reduce the network energy consumption and effectively improve the efficiency and accuracy of data fusion under the same data volume. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
