A Location History-Aware Retail Product Recommender System

Abstract

Mobile Context-Aware Recommender Systems (CARS) have recently emerged aiming at generating recommendations relevant to the specific environmental and situational usage context. This article reports on the design and implementation of a collaborative filtering-based mobile CARS, developed as part of an e-platform that supports location-based search for retail products and services sold by nearby physical retailer shops. In addition to the current user location, our RS considers a multitude of contextual factors like time, season, demographic data, consumer behavior, and location history of the user in order to derive more meaningful product recommendations. The RS has been tested on real operational environments as well as on lab evaluation trials demonstrating higher accuracy and relevance of recommendations against two baseline approaches.