题　目：Generative AI with Guided Embedding Modifier (GEM) Improving Package Design Using Eye Movements

嘉　宾：Jingling Yu（俞晶铃） , Ph.D. Candidate, The Hong Kong University of Science and Technology

主持人：张铄 副教授 awc777万象城娱乐官网

时　间：2024年11月13日（周三）13:30-15:00

地　点：awc777万象城娱乐官网 徐汇校区 awc777万象城娱乐官网A305

内容简介:

Package design is an important marketing mix element that plays a critical role when consumers visually search for products. Whereas package design is a highly subjective and complex process, it remains difficult for firms to quantitatively assess the effectiveness of designs, and consequently, consumers often struggle to find their desired product in cluttered environments, especially under time pressure. Our research aims to develop a methodology that assists managers in enhancing package design to improve product findability while maintaining aesthetic appeal. Our contributions are threefold: 1) We extend the Generative AI framework with a Guided Embedding Modifier (GEM), implementing a interpretable machine-learning tool for encoding and generating product packaging. We also construct a structural model within this framework that connects encoded package design variables to shoppers' eye movements during product searches on e-commerce websites. This approach allows us to understand how packaging design affects visual attention processes during search, which we use to guided improve package design. 2) We explore the moderating impact of time pressure and find that text design on the packaging is especially important, implying that managers should focus on the brand identification part for the time pressure consumers. 3) Our methodology and follow-up valid experiment provide firms valuable guidance in creating attention-grabbing packaging grounded in objective data.

演讲人简介：

Jingling Yu is a Ph.D. Candidate in Marketing from the Hong Kong University of Science and Technology. She received her MPhil degree in marketing from the Hong Kong University of Science and Technology in 2021 and her bachelor’s degree in chemical engineering from Xiamen University in 2019. Jingling is an applied methodologist with research interests at the intersection of computer science and marketing. Her research focuses on visual marketing design using eye-tracking technology and influencer marketing. She also explores how rich, unstructured data—such as images and videos—can be utilized in traditionally challenging areas like package design and influencer content creation.

欢迎广大师生参加!