To solve the problem that the traditional retail market is limited to single-dimensional data analysis of sales when predicting commodity sales, which ignores the long-term trend, seasonality and holiday within the sales data, making it difficult to fully capture the data characteristics and accurately predict sales. This paper constructs a Prophet-LightGBM combined machine learning model. Firstly, the Prophet model is used to automatically decompose data characteristics and flexibly adjust their impact on sales. Then, the LightGBM model is used to build an efficient prediction model for the multi-dimensional features identified by Prophet to improve the reliability of sales forecasts. Experiments have shown that the prediction R2 of the combined model has reached 0.85, which is better than the prediction effect of a single model. It is proved that the proposed combined model has better performance in commodity sales forecasting, and provides strong theoretical data support for enterprises to accurately grasp market demand, optimize production planning and inventory management.

2025-03-31·Frontiers in Artificial Intelligence and Applications

The Design of Personalized Cultivation Program for Enterprise Digital Talents Based on Smart Learning Platform

作者: Gao, Jie ; Zhou, Zhaofeng

In order to solve the problem of low efficiency and quality of enterprise software engineering talent cultivation, the design of personalized cultivation program for enterprise digital talents based on intelligent learning platform is proposed. The design adopts a layered design method, combined with artificial intelligence algorithms, and develops a platform containing functional modules such as course management, student learning tracking, interaction and feedback, data analysis and reporting. In order to verify the effectiveness and performance of the intelligent platform, professional testing software is used to test the functionality and performance of the platform and the traditional platform. The experimental results show that the functional coverage of the intelligentized platform reaches 98.0% (higher than the 92.0% of the traditional platform), and the error discovery rate is 0.5% (lower than the 2.3% of the traditional platform), which indicates that it has higher stability and reliability. In terms of test pass rate, the platform is 99.5% compared to 96.0% for traditional platforms, reflecting the high usability of the platform. In the performance test, under the simulated scenario of 1,000 concurrent users, the average response time of the platform is 1.2s, while the traditional platform is 2.5s, reflecting the significant performance advantage of the platform. In terms of system throughput, the platform can handle 1500 requests per second, which is higher than the traditional platform. In terms of resource utilization, the utilization rates of the CPU and memory of the platform under high load are 70.0% and 65.0% respectively, while those of the traditional platform are as high as 85.0% and 80.0% respectively, indicating that the platform is more efficient in resource management. Conclusion: The intelligent platform outperforms the traditional platform in terms of function coverage, error detection rate, system throughput and other key indexes, which verifies its effectiveness and performance advantages in the modern education environment.

2025-01-17·2025 IEEE 5th International Conference on Power, Electronics and Computer Applications (ICPECA)

Research on the Detection Method of Cigarette Appearance Defect Based on Deep Learning