Enhancing voting advice applications with dynamic profiles

Abstract:

Recommender systems are computer-based techniques mainly used on electronic business to reduce information overload and to provide suggestions of products likely to interest a user. This paper presents an ongoing research of recommender systems applied on eGovernment, particularly it is an extension of so-called voting advice applications (VAAs). Traditional VAAs provide recommendations of political parties and candidates focusing on static profiles of users, while the system architecture presented in this work proposes an advanced approach for extending users' profile to a dynamic state. This paper introduces the use of fuzzy profiles that include both, static and dynamic components. The dynamic profile generation contains different elements such as, context-aware information and privacy and trust concerns of users in order to provide different types of output recommendations and visualizations. Then, the system architecture and a prototype implementation of the extended VAA platform are presented. Furthermore, conclusions and outlook for future work will be discussed.

Año de publicación:

2016

Keywords:

• Voting advice applications
• recommender systems
• privacy
• Dynamic profiles
• Context-awareness
• trust

Fuente:

scopus