This is an extract from “AI for business“, the second report of the four-part series, “Asia’s AI agenda”, by MIT Technology Review Insights.
AI is driving increasingly sophisticated personalization at Chinese e-commerce company Alibaba. Binqiang Zhao, head of personalization, describes the machine learning journey moving from linear regression towards deep and reinforcement learning. By identifying a customer’s long-term preferences on types of items, price, brand, and manufacturer location, Zhao and his team have built “a very deep model that accurately predicts the click and conversion rate.” Alibaba’s app also collects information that proves valuable. For instance, the orientation of the handset, how frequently it is activated, and other indicators signal when a customer is searching for a better price or indicating buying behavior. Since 2016 Alibaba has made a click and conversion improvement of more than 10% in search, and more than 20% in recommendations.