A Carbon Emission Public Opinion Monitoring Platform Based on Big Data Mining Algorithm and Reinforcement Learning Algorithm

Due to the characteristics of large scale and diverse types of carbon emission public opinion data, monitoring is difficult. Therefore, this paper proposes a research on the design of a carbon emission public opinion monitoring platform based on big data mining algorithms and reinforcement learning algorithms. The SOM-3588 high-performance core board is used as the hardware device of the design monitoring platform to provide powerful computing power for carbon emission public opinion data. Using big data mining algorithms and combining the attribute characteristics of the data, a carbon emission public opinion data warehouse was constructed; in the process of carrying out specific public opinion monitoring, a reinforcement learning algorithm was introduced, and the Transformer model based on the full attention mechanism was used to analyze the carbon emission public opinion data. Corresponding opinions and attitudes. In the test results, the accuracy of monitoring results for various types of carbon emissions public opinion sentiments is always above 90.0%. Compared with the control group, it has obvious advantages in reliability and stability respectively.