2026年第34期(总第1178期)
演讲主题：Identifying Purchase-Evoking Social Media Posts: A Theory-Driven Deep Multimodal Learning Framework
主讲人：郑兴 佛罗里达大学教授
主持人：王秭移 计算金融系讲师
活动时间：2026年5月21日（周四）10:00-12:00
活动地址：管院大楼105教室
主讲人简介：
郑兴教授现任美国佛罗里达大学沃灵顿商学院终身教授及John B. Higdon杰出讲座教授，现任信息系统与运营管理系主任。其研究聚焦数字技术在金融中的应用、电子商务、互联网技术对软件开发与营销的影响及信息系统政策。
郑教授在软件开发策略与网络中立性领域的研究具有开创性意义，尤其关于软件盗版原因的研究对企业伦理管理产生重要影响。其论文“Toward a Profile of Student Software Piraters”入选《Journal of Business Ethics》创刊30周年最优秀的49篇论文之一。“The Debate on Net Neutrality—A Policy Perspective”被近152,000个网站引用，并被《商业周刊》报道、接受《计算机世界》采访；“Estimating Social Influences from Social Networking Sites”获《Decision Sciences Journal》最卓越贡献论文奖。根据AIS统计，其在信息系统三大顶级期刊发表数量排名中，2010–2012年位列全球第16位，2016–2020年位列第32位。
郑教授曾任《Information Systems Research》副主编，现任《Decision Sciences Journal》和《Journal of the Association for Information Systems》副主编及资深主编，并担任《Production and Operations Management》特刊主编，亦曾在INFORMS等国际会议担任重要学术职务。
活动简介：
Social media plays an increasingly important role in shaping consumer purchase behavior and informing operational decisions. However, its multimodal nature, combining text and images, makes it difficult for firms to extract actionable insights. To address this challenge, we propose a theory-driven Quality-Credibility-Complementarity Deep Multimodal Learning (QCC-DML) framework. Specifically, we extend the traditional Information Adoption Model (IAM), originally developed for text-based content, to interpret multimodal social media posts. By incorporating content complementarity between text and images, post quality, and source credibility, the framework identifies posts that evoke consumer purchase intentions. Based on these posts, we construct purchase-evoking frequency (PEF), defined as the number of purchase-evoking posts mentioning a specific product feature. Using social media and sales data from a kitchenware manufacturer, we show that PEF has a significant positive effect on product sales. PEF captures feature-level demand signals and reflects which product features are gaining traction among consumers. This effect is stronger for search goods (vs. experience goods) and online sales (vs. offline sales). This study contributes to operations management literature by demonstrating how multimodal social media analytics supports data-driven decision-making. Practically, the framework helps firms detect evolving consumer preferences and improve decisions in product design, inventory planning, and supply chain coordination.
