深度学习法在收入预测问题中的应用

为了提升电信行业收入预测问题准确率,建立基于循环神经网络和长短时记忆网络相结合的收入预测模型。首先对数据作预处理,然后建立卷积层进行核心预测算法优化,再通过训练寻找最优参数,并将其应用于电信运营商收入预测。实验结果表明,该模型可以预测出未来一个月或者几个月的收入增减变化趋势,预测准确率比传统方法提高20%,算法收敛性也提高约15%。该模型预测结果对于电信行业制定营销方案具有较好指导作用。

Application of Deep Learning Method in Revenue Forecast

In order to improve the accuracy of revenue forecast in communication industry, the arevenue forecast model based on long short-term memory network of recurrent neural network is established in this paper. First, the data is preprocessed, then the convolutional layer is established to optimize the core prediction algorithm. The training is carried out to find the optimal parameters, and it is applied to the revenue prediction of telecom operators. The experimental results show that the model can predict the overall increase and decrease trend of income in the next month or several months. The prediction accuracy is improved by 20% compared with the traditional method, and the convergence of the algorithm is improved by about 15%. The prediction results of the model have a good guiding role for the communication industry to develop sales plans.