From an analytical, conceptual and technical point of view, the project investigates the implementation of a situation-specific recommendation system that adapts to the customer's stress level and thus his susceptibility to product recommendations.
Motivation:
The booming e-commerce and the advancing digitalization demand stationary trade. This is why convincing services are needed to attract customers to stores. A smart service that has gained in importance in recent years, especially in e-commerce, and has found its way into stationary retail, is the recommender system or mobile recommendation system. Here, the customer is given product recommendations based on previous shopping history. However, the shopping experience or the shopping situation in the shop itself differs from that in the online shop. For example, the potential for increased stress is very high due to physical and mental exertion during shopping, overcrowded stores, long queues at the checkout and the fact that customers have little time to shop. Previous approaches neglected the receptivity or mood of the customer during the shopping process in order to optimize the placement of product recommendations on mobile devices.
Objective:
From an analytical, conceptual and technical point of view, the project investigates the implementation of a situation-specific recommendation system that adapts to the customer's stress level and thus his susceptibility to product recommendations. One focus of this project is, therefore, the analysis of customer behavior during shopping from a neuroscientific perspective to analyze the mood/stress level and susceptibility of the customer and thus give product recommendations if the customer is receptive to it. In addition, the product recommendation data will be stored locally in a closed system on the customer's mobile device and analyzed using machine learning methods so that the customer retains full control over his personal data.
Partners
Media-Saturn
