使用Keras构建的神经网络模型对电信网络欺诈识别研究  

Research on Fraud Recognition in Telecommunication Network Using Neural Network Model Constructed by Keras

作　　者：段雪莹[1] 孟小泸 DUAN Xueying;MENG Xiaolu(Jilin Police College,Changchun Jilin 130017;Changchun Sci-Tec University,Changchun Jilin 130600)

机构地区：[1]吉林警察学院,吉林长春130017 [2]长春科技学院,吉林长春130600

出　　处：《软件》2024年第7期95-97,104,共4页Software

基　　金：吉林警察学院院级科研项目“基于机器学习的电信网络诈骗行为识别研究”(jykyzd202404)

摘　　要：本研究旨在探究电信网络欺诈识别中深度学习模型的应用。采用基于Keras构建的神经网络模型,对欺诈数据集进行了详细的数据准备和模型设计,并对模型训练方法进行了详细阐述。实验结果表明,本文设计的神经网络模型相较于基准模型,在准确率、精确率、召回率和F1分数等评估指标上均有显著提升,尤其是在处理不平衡数据集时表现更为优异。本研究为电信网络欺诈识别提供了一种有效的深度学习解决方案,为相关领域的进一步研究和实践提供了借鉴。This study aims to explore the application of deep learning model in telecommunication network fraud recognition.Using the neural network model based on Keras,the fraud data set is prepared and the model is designed in detail,and the model training method is described in detail.The experimental results show that compared with the benchmark model,the neural network model designed in this paper has significantly improved the accuracy rate,accuracy rate,recall rate,F1 score and other evaluation indicators,especially when dealing with unbalanced data sets.This study provides an effective deep learning solution for telecommunication network fraud identification,and provides reference for further research and practice in related fields.

关 键 词：电信网络欺诈识别 神经网络 Keras框架 

分 类 号：TN915.08[电子电信—通信与信息系统]