针对当前专利技术分成率评估方法出现的问题,引入BP神经网络方法来计算专利技术分成率。第一步是建立指标体系。第二步是训练样本的选取以及数据标准化处理。第三步是完成模型误差调优。最后,通过实例验证了使用BP神经网络方法评估专利技术分成率是可行的。 Aiming at the problems that occur in the current assessment method of patent technology share rate, BP neural network method is introduced to calculate the patent technology share rate. The first step is to establish an indicator system. The second step is the selection of training samples and data standardization. The third step is to complete the model error tuning. Finally, an example is given to verify that it is feasible to use BP neural network to evaluate the share of patent technology.