Applying transfer learning to achieve precision marketing in an omni-channel system – a case study of a sharing kitchen platform
研究开发了一个整合iOS、Android和Web组件的全渠道聊天机器人，利用卷积神经网络和迁移学习实现个性化服务和精准营销，并通过共享厨房案例验证其可迁移性。
Omni-channel marketing is an enhanced cross-channel business model involving shared data that allows enterprises to enhance and facilitate customer experience. Omni-channel opportunities shape retail business and shopper behaviours by coordinating data across all channel platforms while enabling their simultaneous use. Artificial intelligence (AI) has played an increasingly critical role in marketing analysis. With the proper training, AI can predict consumer preferences and provide recommendations based on historical data to achieve precision marketing in e-commerce. At present, however, the existent chatbots on many product-ordering platforms lack AI refinement, resulting in the need to ask customers multiple questions before generating a reliable suggestion, yet an effective way to incorporate AI in an omni-channel platform has remained vague. Hence, the aim of this study was to develop an omni-channel chatbot that incorporates iOS, Android, and web components. The chatbot was designed to achieve personalised service and precision marketing using convolutional neural networks (CNNs). A shared kitchen case study demonstrates the advantages of the proposed method, which is transferable to other consumer applications such as clothing selection or personalised services. The number of food offerings and the quality of image classifiers set the research limitations, pointing toward the direction of future research.