基于Aspect-Based LSTM的贵州刺梨电商评论细粒度情感分析
Fine-Grained Sentiment Analysis of Guizhou Rosa roxburghii E-Commerce Reviews Based on Aspect-Based LSTM
摘要: 为精准挖掘用户对特色农产品的需求偏好，助力电商经济高质量发展，本文基于Aspect-Based LSTM模型，对京东6045条贵州刺梨相关评论展开细粒度情感分析。经数据清洗与标注，构建6505条“评论、方面及情感”三元样本，实验表明模型准确率达97%，宏平均F1值为0.92，加权平均F1值为0.97，核心维度识别准确率超95%，数据可靠性高。本研究通过对模型数据的深度挖掘，提炼出用户对产品品质、使用体验与健康价值的核心需求。研究发现，物流防护不足、产品标准化缺失与价值传递效率低下是制约其发展的三个关键因素。据此本文从产品、渠道、营销端三方面系统性地提出电商运营优化建议，为贵州刺梨及同类特色农产品的电商推广提供数据支撑，以赋能产业发展，助推乡村振兴。
Abstract: To accurately identify user preferences for specialty agricultural products and promote high-quality development of the e-commerce economy, this paper, using an Aspect-Based LSTM model, conducted a fine-grained sentiment analysis of 6045 reviews related to Guizhou Rosa roxburghii on JD.com. After data cleaning and labeling, 6505 triplet samples of “review, aspect, and sentiment” were constructed. Experiments show that the accuracy of the model is 97%, the macro average F1 value is 0.92, the weighted average F1 value is 0.97, the core dimension recognition accuracy is more than 95%, demonstrating high data reliability. Through in-depth mining of the model data, this paper identified core user needs for product quality, user experience, and health benefits. The paper found that inadequate logistics protection, lack of product standardization, and inefficient value delivery were three key factors hindering its development. Therefore, this paper systematically proposes e-commerce operational optimization recommendations from the product, channel, and marketing perspectives. This provides data support for the e-commerce promotion of Guizhou Rosa roxburghii and similar specialty agricultural products, empowering industrial development and promoting rural revitalization.
文章引用：许祖娟, 夏雨欣, 王兴隆, 张汉林. 基于Aspect-Based LSTM的贵州刺梨电商评论细粒度情感分析[J]. 电子商务评论, 2025, 14(11): 2341-2353. https://doi.org/10.12677/ecl.2025.14113696

参考文献

[1] 安华明, 刘明, 杨曼, 等. 刺梨有机酸组分及抗坏血酸含量分析[J]. 中国农业科学, 2011, 44(10): 2094-2100.
[2] 龚映梅, 李紫荆, 王余洲. 基于在线评论的生鲜电商平台物流服务质量评价比较[J/OL]. 昆明理工大学学报(自然科学版): 1-10. 2025-09-18. [Google Scholar] [CrossRef] 
[3] 王俊. 基于BERT与注意力机制在陶瓷电商产品评论中的情感分析应用[D]: [硕士学位论文]. 景德镇: 景德镇陶瓷大学, 2025.
[4] Mahalakshmi, S. and Sivasankar, E. (2015) Cross Domain Sentiment Analysis Using Different Machine Learning Techniques. In: Ravi, V., Panigrahi, B., Das, S. and Suganthan, P., Eds., Proceedings of the Fifth International Conference on Fuzzy and Neuro Computing (FANCCO—2015), Springer, 77-87. [Google Scholar] [CrossRef] 
[5] Pontiki, M., Galanis, D., Papageorgiou, H., Androutsopoulos, I., Manandhar, S., AL-Smadi, M., et al. (2016) Semeval-2016 Task 5: Aspect Based Sentiment Analysis. Proceedings of the 10th International Workshop on Semantic Evaluation (SemEval-2016), San Diego, 16-17 June 2016, 19-30. [Google Scholar] [CrossRef] 
[6] Hu, M. and Liu, B. (2004) Mining and Summarizing Customer Reviews. Proceedings of the Tenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Seattle, 22-25 August 2004, 168-177. [Google Scholar] [CrossRef] 
[7] Tang, D., Qin, B. and Liu, T. (2016) Aspect Level Sentiment Classification with Deep Memory Network. Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, Austin, 1-4 November 2016, 214-224. [Google Scholar] [CrossRef] 
[8] Li, H., Yu, B.X.B., Li, G. and Gao, H. (2023) Restaurant Survival Prediction Using Customer-Generated Content: An Aspect-Based Sentiment Analysis of Online Reviews. Tourism Management, 96, Article ID: 104707. [Google Scholar] [CrossRef] 
[9] Mehra, P. (2023) Unexpected Surprise: Emotion Analysis and Aspect Based Sentiment Analysis (ABSA) of User Generated Comments to Study Behavioral Intentions of Tourists. Tourism Management Perspectives, 45, Article ID: 101063. [Google Scholar] [CrossRef] 
[10] Graves, A. (2012) Long Short-Term Memory. In: Graves, A., Ed., Supervised Sequence Labelling with Recurrent Neural Networks, Springer, 37-45. [Google Scholar] [CrossRef] 

为你推荐