The ShophIA project aims, firstly, to improve in-store and online data gathering through massive data sources, thereby increasing the granularity of the information collected and generated and, secondly, to investigate in federated deep symbolic learning algorithms sustained by an edge computing architecture that will improve the customer experience with a virtual assistant both in the shop and on the web.
Regarding the first line of action, related to data gathering, the ShophIA project will employ the following two main mechanisms:
- In-store, through a modular hardware solution (blackbox) which, based on a set of devices (cameras, digital panels, IoT -iBeacons, Smart NFC Tags, etc.) will enable the generation of knowledge through tools based on artificial intelligence (pattern detection, target tracking, image recognition…), providing critical information for internal management of stock and sales.
- On-line (application and social networks) to generate personalised campaigns (omnichannel marketing) capable of identifying the needs of consumers, maximising the commercial objectives of clients. Regarding the second line of action, related to data processing, the ShophIA project will investigate in the following main areas.
- Federated edge computing, which will process data collected in-store to provide personalised assistance and collect information during the shopping experience (e.g., by performing sentiment analysis through facial recognition), to increase customer knowledge and improve customer relationships.
- Processing in the Cloud with Deep Symbolic Learning techniques, to learn the preferences of consumers and parameters that most influence their purchasing decisions, analyse the sentiment produced by the brand or manage the loyalty of new customers.
The consortium formed by the company Eurostar Mediagroup and the technology centre AIR Institute, proposes research into a functional model of intelligent system for the retail sector based on an edge computing platform. The result of the ShophIA project will be an open, flexible and scalable solution that allows for the comprehensive management of the company’s relationship with the customer. To this end, it will merge sources in a knowledge base based on the online and in-store profiles to generate a personalised assistance experience (retail assistant) through the functional model and the equipment deployed in-store. Finally, ShophIA will collect customer feedback and carry out a followup to assess customer satisfaction.
