题目： Learning consumer preference structures for purchase decisions based on online reviews
报告人：吴性丽
报告时间：2023年12月20日（周三）19:00-21:30
交流平台：腾讯会议：883-772-885
报告人简介：
吴性丽，四川大学管理学博士，现为四川大学特聘副研究员。研究方向为决策分析、在线评论、偏好学习等。在Omega、European Journal of Operational Research、Technological Forecasting and Social Change、Information Processing & Management、IEEE Transactions on Fuzzy Systems等海外高水平期刊上以第一/通讯作者身份发表学术论文20余篇，其中，5篇入选ESI高被引论文，3篇入选ESI热点论文。入选斯坦福大学发布的全球前2%科学家榜单（2020-2023）。主持国家自然科学基金青年项目、国家社科后期优秀博士论文出版项目各一项。担任海外高水平期刊《Journal of Intelligent and Fuzzy Systems》副主编。
报告摘要：
Accurate prediction of consumer preferences aids numerous marketing efforts in e-commerce platforms, as it describes which product features impact and how they impact a consumer’s purchase decisions. We propose a preference learning approach for automated extraction of consumer preferences from online reviews. A preference model based on the multi-attribute value theory is proposed, taking into account attribute importance, attribute interactions, consumer risk tolerance, and consumer judgment benchmarks. To estimate parameters in the preference model, we develop a regression model based on the non-linear relationship between the overall utility determined through product ratings and attribute-level utilities derived from text reviews.
| 上一条： 没有了 |
| 下一条： 没有了 |
