面向平台服务优化的电子商务客户评论多语义识别研究
Research on Multi-Semantic Identification of E-Commerce Customer Reviews for Platform Service Optimization
摘要: 电子商务客户评论反映了消费者对商品质量、价格、物流、售后等方面的真实感受，是平台优化服务机制和商家改进运营管理的重要依据。然而，客户评论多语义标签因标注成本高、表达复杂等而普遍存在标签缺失问题。基于此，文章围绕电子商务客户评论的信息挖掘需求，构建客户评论多语义识别框架，研究标签缺失条件下的评论主题恢复问题。具体地，结合缺失多标签学习思想引入精确秩函数刻画评论标签空间中的相关性，并采用PALM算法进行求解。实验结果表明，所提方法在电子商务评论数据上具有较好的识别效果与稳定性，能够更有效地提取消费者对价格、物流、满意程度等方面的信息。研究结果可改进商家服务质量以及为平台治理提供参考。
Abstract: E-commerce customer reviews reflect consumers’ genuine perceptions of product quality, price, logistics, after-sales service, and other aspects, and thus provide an important basis for platform service optimization and merchant operation improvement. However, multi-semantic labels in customer reviews are often incomplete due to the high cost of annotation and the complexity of review expressions. To address this issue, this paper focuses on the information mining needs of e-commerce customer reviews, constructs a multi-semantic identification framework for customer reviews, and investigates the problem of review topic recovery under missing-label conditions. Specifically, drawing on the idea of incomplete multi-label learning, this study introduces an exact rank function to characterize the correlations within the review label space and employs the PALM algorithm for optimization. Experimental results show that the proposed method achieves fairly good identification performance and stability on e-commerce review datasets, and can more effectively extract information related to price, logistics, and satisfaction. The findings provide useful support for improving merchant service quality and informing platform governance.
文章引用：朱文娇, 彭定涛. 面向平台服务优化的电子商务客户评论多语义识别研究[J]. 电子商务评论, 2026, 15(10): 524-533. https://doi.org/10.12677/ecl.2026.15101148

参考文献

[1] Jia, Q.W., Deng, T.Q., Yang, M., Wang, Y. and Wang, C.Z. (2025) Class Label Fusion Guided Correlation Learning for Incomplete Multi-Label Classification. Information Fusion, 120, Article 103072.
https://doi.org/10.1016/j.inffus.2025.103072
[2] Huang, J., Qin, F., Zheng, X., Cheng, Z.K., Yuan, Z.X., Zhang, W.G. and Huang, Q.M. (2019) Improving Multi-Label Classification with Missing Labels by Learning Label-Specific Features. Information Sciences, 492, 124-146.
https://doi.org/10.1016/j.ins.2019.04.021
[3] Zhu, Y., Kwok, J.T. and Zhou, Z. (2018) Multi-Label Learning with Global and Local Label Correlation. IEEE Transactions on Knowledge and Data Engineering, 30, 1081-1094.
https://doi.org/10.1109/tkde.2017.2785795
[4] Pan, H., Xu, H., Zheng, J., Shao, H. and Tong, J. (2024) A Semi-Supervised Matrixized Graph Embedding Machine for Roller Bearing Fault Diagnosis under Few-Labeled Samples. IEEE Transactions on Industrial Informatics, 20, 854-863.
https://doi.org/10.1109/tii.2023.3265525
[5] Zhang, H., Qian, J., Zhang, B., Yang, J., Gong, C. and Wei, Y. (2020) Low-Rank Matrix Recovery via Modified Schatten-P Norm Minimization with Convergence Guarantees. IEEE Transactions on Image Processing, 29, 3132-3142.
https://doi.org/10.1109/tip.2019.2957925
[6] Zhang, S. and Xin, J. (2018) Minimization of Transformed L1 Penalty: Theory, Difference of Convex Function Algorithm, and Robust Application in Compressed Sensing. Mathematical Programming, 169, 307-336.
https://doi.org/10.1007/s10107-018-1236-x
[7] Bolte, J., Sabach, S. and Teboulle, M. (2014) Proximal Alternating Linearized Minimization for Nonconvex and Nonsmooth Problems. Mathematical Programming, 146, 459-494.
https://doi.org/10.1007/s10107-013-0701-9
[8] Horn, R.A. and Johnson, C.R. (1985) Matrix Analysis. Cambridge University Press.
https://doi.org/10.1017/cbo9780511810817
[9] Deniz, E., Erbay, H. and Coşar, M. (2022) Multi-label Classification of E-Commerce Customer Reviews via Machine Learning. Axioms, 11, Article 436.
https://doi.org/10.3390/axioms11090436

为你推荐