A privacy-aware feature selection method for solving the personalization–privacy paradox in mobile wellness healthcare services

• The essence of personalization in mobile wellness healthcare (MWH) services is feature selection.
• A novel method of privacy-aware feature selection for individualized MWH is presented.
• An optimal set of features is automatically found to solve the personalization–privacy paradox.
• The results successfully show that the proposed method outperforms conventional methods.

Despite the vast number of available opportunities, customers’ privacy concerns can inhibit their acceptance of mobile wellness healthcare services. Personalization of these services may go some way toward alleviating these concerns. The essence of personalization in mobile wellness healthcare services is feature selection because users may be concerned about disclosing private information, although it may be useful in provision of personalized services. Therefore, an optimal feature selection method is needed which considers both these privacy concerns and the quality of personalization. Such safe and accurate data collection would facilitate understanding of customers’ preferences while safeguarding their privacy. In this study, a privacy-aware feature selection method is proposed based on the personalization–privacy paradox, and this paradox is explored in the context of wellness healthcare services with consideration of the personal characteristics of customers using these services.