Implementación de un prototipo de sistema de recomendación para publicidad inteligente

Abstract

This document presents the implementation of a recommender system prototype for smart publicity. The prototype is composed of 4 components: Android client, WCF service, database and a recommender algorithm. The Android client will be responsible for collecting and forwarding device information, such as the location. This information will be stored in the MySQL database in conjunction with the publicity from shops. The Android client will make requests for recommendations through the WCF service. The recommender algorithm will use as inputs the information collected by the Android client, as well as the information of shops, so that they can be processed and returned as recommendations. C# programming language was used to develop both the WCF service and the recommendation algorithm, to solve basic tasks in the recommender system such as the classification of predictions and the categorization of items. Java programming language was used for the development of the Android client.