The deep separable convolution with DSC NCF model and optimization mechanism of digital economy for intelligent manufacturing under sales order recommendation algorithm

基于DSC NCF模型的深度可分离卷积及面向智能制造的数字经济销售订单推荐算法的优化机制

阅读:2

Abstract

This study aims to explore the optimization role of deep learning technology in sales order management for smart manufacturing enterprises within the context of the digital economy, as well as its driving mechanism for industrial structure upgrading and smart transformation. Specifically, the study focuses on how deep learning algorithms can improve the efficiency of order management and customer satisfaction in smart manufacturing enterprises, thereby promoting their intelligent transformation. The study employs the Deep Separable Convolutional Neural Collaborative Filtering (DSC-NCF) algorithm, combined with the publicly available smart manufacturing dataset Alibaba Click and Conversion Prediction (Ali-CCP), to build a deep learning-based intelligent recommendation platform. By comparing it with traditional Neural Collaborative Filtering (NCF), Factorization Machine (FM), and other benchmark algorithms, the study evaluates key performance indicators such as accuracy, recall, F1 score, and Area Under the ROC Curve (AUC) of the DSC-NCF algorithm across different training epochs. The experimental results demonstrate the significant superiority of the DSC-NCF algorithm across all training epochs. Specific data are as follows: (1) Accuracy: At 100 training epochs, DSC-NCF achieves 0.91, significantly higher than NCF (0.87), FM (0.81), and other benchmark algorithms. (2) Recall: At 100 training epochs, DSC-NCF achieves 0.92, outperforming NCF (0.86) and FM (0.80). (3) F1 Score: At 100 training epochs, DSC-NCF achieves 0.94, significantly higher than NCF (0.89) and FM (0.84). (4) AUC: At 100 training epochs, DSC-NCF achieves 0.99, significantly higher than NCF (0.95) and FM (0.91). These results indicate that the DSC-NCF algorithm has significant advantages in feature extraction and model optimization, enabling it to more accurately capture complex relationships between users and items, thereby enhancing the performance of the recommendation system. By accurately predicting customer demands and optimizing order processes, enterprises can better adapt to market changes, improve customer satisfaction, and drive industrial structure upgrading and smart transformation. This study not only provides an efficient deep learning-based order management solution for smart manufacturing enterprises but also offers theoretical support and technical pathways for industrial structure upgrading in the context of the digital economy.

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。