基于多尺度深度学习的B2C电商物流网络信息实时提取

Real-time Extraction of B2C E-commerce Logistics Network Information Based on Multi-scale Deep Learning

传统信息提取方法难以应对大规模的物流数据,导致提取的信息噪声较多。因此,提出基于多尺度深度学习的B2C电商物流网络信息实时提取。挖掘B2C电商物流网络信息,预处理收集到的原始电商物流网络信息。利用多尺度深度学习对预处理后的电商物流信息进行特征重构,多尺度深度学习捕获不同尺度的特征,生成新的特征矢量提取出重构后的电商物流网络信息。实验结果证明,所研究方法在提取电商物流网络信息时,噪声含量较少,提取效果较好。

Traditional information extraction methods are difficult to deal with large-scale logistics data,resulting in more information noise.Therefore,the real-time extraction of B2C c-commerce logistics network information based on multi-scale deep learning is proposed.Mining B2C c-commerce logistics nctwork information and preprocessing the collected original c-commerce logistics network information.Multi-scale deep learning is used to reconstruct the features of pre-processed e-commerce logistics information.Multi-scale deep learning captures the features of different scales and generates new feature vectors to extract the reconstructed e-commerce logistics nctwork information.The experimental results show that the research method has less noise content and better extraction effect when extracting c-commerce logistics network information.