Intelligent Asset Allocation Portfolio Division and Recommendation: Based on Deep Learning and Knowledge Graphs

With the continuous development of financial markets, intelligent asset allocation has become a topic of great concern in the investment field. However, traditional asset allocation methods often face difficulties in grasping the relationship between diversity, risk and return, which limits its application in complex market environments. To solve this problem, this study introduces deep learning and knowledge graphs and proposes an intelligent asset allocation model. Our model makes full use of the advantages of the Knowledge Graph Embedding Model (KGE), LSTM, and Genetic Algorithm (GA) to build a multi-level and multi-dimensional asset allocation model. KGE helps capture the complex relationships between different assets, LSTM is used to learn key patterns of historical portfolio performance, and GA finds the optimal asset allocation combination by simulating natural selection and genetic mechanisms. Experimental findings indicate that our model has demonstrated substantial improvements across various performance metrics and outperforms conventional approaches.