喻园管理论坛2019年第27期(总第478期)
演讲主题：Big Data and Machine Learning: The Paradigm Change in Marketing
主 讲 人：邱淳(Martin Qiu)，加拿大劳埃尔大学拉扎里得斯商学院营销系副教授(终身教职)
主 持 人：阎俊，威尼斯欢乐娱人城·首页工商管理系副教授
活动地点：威尼斯欢乐娱人城·首页412教室
活动时间：2019年4月18日（周一）19:00-20:30
主讲人简介：
邱淳博士于1996年本科毕业于威尼斯欢乐娱人城·首页经济学院，2008年获得加拿大阿尔伯塔大学（University of Alberta）商学院营销专业博士学位，曾任教于加拿大麦基尔大学McGill University，现为加拿大劳埃尔大学拉扎里得斯商学院（Lazaridis School of Business and Economics, Wilfrid Laurier University）营销系副教授(终身教职)。他的研究方向是零售管理，大数据营销，移动营销和企业社会责任，他的研究成果发表在Journal of Retailing，European Journal of Marketing， Marketing Letters，Journal of Economic Psychology, Journal of Business Ethics等一流商业期刊上。
讲座简介：
Rapid advances in information technologies has made today’s marketers realize that experience and intuition do not work as effectively as before. They start to embrace data driven decisions, in which big data and machine learning play a pivotal role hand in hand. This talk introduces the current paradigm change in marketing field, and covers major machine learning algorithms that have wide applications in marketing, including market basket analysis, recommender systems, and RFM analysis.
Market basket analysis is a data mining technique commonly used in retail business to understand the purchase behavior of the consumer – what groups of items do consumers tend to purchase together? As an obvious example, such analysis might reveal that people tend to buy shampoo and conditioner together. This information can then be used to create appropriate sales promotions or the placement of the products in a supermarket.
Recommender systems aim to predict users’ interests based on certain information about those users, and recommend items that quite likely are interesting for them. They are utilized in a variety of marketing fields including recommendations for entertainment products such as movies and music, as well as e-commerce websites for shoppers.
RFM analysis examines how recently customers have purchased (recency), how often they purchase (frequency), and how much they spend (monetary), and assigns value to them based on their RFM pattern in order to segment them, and target them later with different CRM programs.
