Research on the business performance evaluation method for small and medium-sized enterprises in cross-border e-commerce based on artificial bee colony optimized LSTM model

基于人工蜂群优化LSTM模型的跨境电商中小企业业务绩效评价方法研究

阅读:1

Abstract

The rapid development of the cross-border e-commerce industry has posed challenges for small and medium-sized enterprises (SMEs), such as large fluctuations in the international market environment and limited resource allocation, which increases the uncertainty and complexity of their business performance. Traditional business performance evaluation methods are inadequate in handling complex nonlinear data and real-time responsiveness, making them difficult to meet the demands of a dynamic market environment. Based on data from 10 cross-border e-commerce SMEs from 2020 to October 2023, this paper proposes a business performance evaluation method based on the Artificial Bee Colony Optimized Long Short-Term Memory Neural Network (ABC-LSTM) to improve the predictive accuracy of complex time series data analysis and the model's generalization ability. The ABC-LSTM model outperforms GA-LSTM, XGBoost, and traditional LSTM models in performance metrics such as Mean Squared Error (0.037), Mean Absolute Error (0.016), and Time Dependency Error (0.019), demonstrating faster convergence speed and higher stability. Additionally, this study analyzes the hierarchical characteristics of performance among different enterprises, revealing the advantages of high-performance enterprises in resource integration, supply chain management, and market expansion, as well as the bottleneck issues of low-performance enterprises. The results validate the significant advantages of the ABC-LSTM model in evaluating the business performance of SMEs in cross-border e-commerce. It not only improves the accuracy of multi-dimensional business data analysis for cross-border e-commerce enterprises but also provides a scientific basis for enterprises in resource integration, supply chain management, and market expansion.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。