Privacy-Preserving Crowdsourcing-Based Recommender Systems for E-Commerce & Health Services

Identifier:  TDX:2634

Handle: https://hdl.handle.net/20.500.11797/TDX2634

Authors:  Casino Cembellin, Francisco Jose

Abstract:
Our society lives an age where the eagerness for information has resulted in problems such as infobesity, especially after the arrival of Web 2.0. In this context, automatic systems such as recommenders are increasing their relevance, since they help to distinguish noise from useful information. However, recommender systems such as Collaborative Filtering have several limitations such as non-response and privacy. An important part of this thesis is devoted to the development of methodologies to cope with these limitations. In addition to the previously stated research topics, in this dissertation we also focus in the worldwide process of urbanisation that is taking place and the need for more sustainable and liveable cities. In this context, we focus on smart health solutions and efficient wireless channel characterisation methodologies, in order to provide sustainable healthcare in the context of smart cities.