According to the current situation and future trend of e-commerce investment market research at home and abroad, combined with the established research objectives, this paper studies and analyzes the existing methods of investment quality evaluation. This paper puts forward the modeling idea of the evaluation index system of investment decision quality, and takes the deep neural network as the modeling tool and python as the simulation tool to complete the establishment of the investment decision support model of e-commerce listed companies and get the experimental results. In this paper, based on the deep learning portfolio replication algorithm, the features of the portfolio model are extracted by the automatic encoder to realize the portfolio replication. First, the definition of deep learning model and the framework of model training will be given. Then, a process of trestle automatic encoder model will be introduced, and then the detailed process of portfolio replication algorithm based on deep learning will be explained in detail. Finally, the trained model is used to evaluate the samples of the test set, and the results are good, which shows that the index evaluation system is feasible and effective.

根据国内外电子商务投资市场研究的现状和未来趋势，结合既定的研究目标，对现有的投资质量评价方法进行研究和分析。本文提出了投资决策质量评价指标体系的建模思路，以深层神经网络为建模工具，以python为仿真工具完成了电子商务上市公司投资决策支持模型的建立。得到实验结果。本文基于深度学习的投资组合复制算法，通过自动编码器提取投资组合模型的特征，以实现投资组合复制。第一，给出了深度学习模型的定义和模型训练的框架。然后，介绍栈桥自动编码器模型的过程，然后详细说明基于深度学习的组合复制算法的详细过程。最后，使用训练好的模型对测试集样本进行评估，结果良好，表明该指标评估系统是可行和有效的。