| Since the21st century, with e-commerce developing rapidly, it threats the original shopping markets. The traditional offline shopping pattern suffers a serious challenge. Offline shopping pattern is facing the problem that how to get customer’s shopping behavior, increase customer shopping experience so as to meet customer consumption demand.This paper aims to design a complete offline shopping recommendation system to collect customer offline shopping behavior data, provide customers with product recommendations and navigation in order to enhance the customer’s shopping experience.This paper writes how to build customer’s offline shopping model, how to identify offline shopping behavior and how to provide customers with a customized message push.This paper through a detailed analysis of customer shopping behavior, gives the definition of the offine shopping behavior as well as designing a shopping recommendation system. The system uses location-based fingerprint WiFi indoor positioning technology to track customer shopping route, through the hotspot recognition technology dividing the mall into various shopping areas to distinguish the customers’ shopping area, using collaborative filtering technology to provide customers with accurate product recommendations. And use simulating tools to simulate the actual scenes, design test cases, do functional, performance and stress testing, through questionnaires to understand the customer’s shopping experience.By target validation, project can be drawn that the system effectively improves the customer’s shopping experience, enhance the flow of people and goods sales in the mall. The main contribution of this paper are:①gives offline shopping behavior model;②improved irregular hotspot design;③customer recommendation based on historical shopping data. |
