一种产品数据分析方法及系统
By performing text normalization and multi-source sentiment representation fusion on product review data, the problem of the synergistic relationship between rating levels and review content was solved, achieving more accurate sentiment analysis and improving the reliability of product feedback.
CN122415137APending Publication Date: 2026-07-17
View PDF 0 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-05-26
- Publication Date
- 2026-07-17
AI Technical Summary
Technical Problem
Existing product data analysis methods struggle to balance the synergy between rating levels and comment content, resulting in unstable sentiment analysis results that fail to accurately reflect users' true emotions.
Method used
By acquiring comment data, performing text normalization processing, generating multi-source sentiment representations, and inputting them into an ensemble learning classification model for feature fusion, generating collaborative sentiment features, and finally outputting the sentiment category.
Benefits of technology
It improves the accuracy and robustness of sentiment assessment in product reviews, reduces noise interference, and can more effectively identify user attitudes, providing reliable data support for product optimization.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN122415137A_ABST
Abstract
本发明涉及数据分析技术领域，具体涉及一种产品数据分析方法及系统，包括：获取目标产品的评价等级信息、评论标题信息和评论正文信息，对评论标题信息和评论正文信息进行词项切分、传播标记剔除、非语义符号剔除、停用词剔除及词干化处理，得到规范评论文本；分别基于评价等级信息、规范评论文本中的倾向词项以及规范评论文本与预设情绪词汇库的匹配结果，生成第一情感表征、第二情感表征和第三情感表征；将三类情感表征融合为协同情感特征，并输入训练后的集成学习分类模型，输出评论数据对应的情感类别。本发明可以解决电子交易平台难以准确反映用户真实情感的问题。
Need to check novelty before this filing date? Find Prior Art