Research on reducing wildlife trade based on neural network model

The illegal trade of wildlife has been recognized as the fourth largest illegal trade in the world, and the illegal trade of wildlife poses a great threat to ecological balance and biodiversity on a global scale. To alleviate this problem, we first identified our customer segments by creating a word cloud to visualize how often they occur. The results show that the global animal society search is particularly obvious, so this paper chooses WAP as the main client. Then, using WAP as the key word for the second search, the 9 indicators with the highest frequency of WAP are finally derived as the index evaluation system of this problem, and the collected indicators are preliminatively analyzed by using the feature correlation heat map. Then the linear regression model and the neural network are used to plot and calculate the 9 indicators of the two customers, get the weight ratio, and deduce the selected customers. The difference between the predicted and measured values of the model is plotted using the training loss curve and the discrete distribution of the measured and predicted values. The development of this comprehensive model not only helps to clarify what additional powers and resources the client needs, but also provides specific guidance and decision support.