| 摘 要: 随着人工智能和金融科技的快速发展，机器学习尤其是深度学习在金融领域的应用引起了人们浓厚的研究兴趣。为了探索金融深度学习的应用领域，对近十年金融深度学习的文献进行了总结，并分别从模型介绍和应用领域两个方面进行了归纳。结果发现：金融深度学习常用的模型包括卷积神经网络、循环神经网络和长短期记忆神经网络，并且它们在金融文本分析、金融风险评估和异常检测以及投资组合管理方面均有着广泛的应用。未来可以将新的文本挖掘和自然语言处理技术应用到行为金融学领域进行更深入的研究，同时还可以探索将深度学习应用到加密货币和区块链等新兴金融领域的更多可能性。 |
| 关键词: 深度学习；神经网络；文本分析；风险评估；投资组合管理 |
|
中图分类号: TP399
文献标识码: A
|
|
| A Research Overview of the Application of Deep Learning in the Financial Field |
|
FU Yufei, WANG Mingyan
|
(School of Management, Shanghai University of Engineering Science, Shanghai 201620, China)
291732841@qq.com; wmy61610@126.com
|
| Abstract: With the rapid development of artificial intelligence and financial technology, the application of machine learning, especially deep learning in the financial field, has aroused strong research interest. In order to explore the application fields of financial deep learning, this paper proposes to summarize the literature of financial deep learning over the past ten years, from two separate aspects: model introduction and application field. Results show that the commonly used models of financial deep learning include convolutional neural network, recurrent neural network and long short-term memory neural network. They have a wide range of applications in financial text analysis, financial risk assessment and anomaly detection, and portfolio management. In future, new text mining and natural language processing techniques can be applied to the field of behavioral finance for more in-depth research. At the same time, more possibilities for applying deep learning to emerging financial fields such as cryptocurrency and blockchain can also be explored. |
| Keywords: deep learning; neural network; text analysis; risk assessment; portfolio management |
