Phone number fraud mode identification method based on graph network and machine learning

The invention belongs to the technical field of communication information security application, and particularly relates to a phone number fraud mode recognition method based on a graph network and machine learning. Compared with the prior art, the telephone number fraud mode recognition method based on the graph network and the machine learning has the advantages that the graph network is firstly used for performing preliminary analysis, the calculation amount of samples is reduced, and the accuracy of the samples is improved; and a corresponding fraud model is obtained through a machine learning algorithm, so that the fraud number identification accuracy is improved. And a favorable escort is provided for communication security and harmony. 本发明属于通讯信息安全应用技术领域，尤其涉及一种基于图网络和机器学习的电话号码诈骗模式识别的方法。与现有技术相比，本发明的优点和积极效果在于，本发明提供一种基于图网络和机器学习的电话号码诈骗模式识别的方法，首先利用图网络进行初步分析，减少样本的计算量以及提高样本的准确率，在通过机器学习算法得到相应的诈骗模型，从而提高诈骗号码识别的准确率。为通信安全、和谐提供有利的护航。