Privacy Risk Perception of Online Medical Community Users Based on Deep Neural Network

Abstract

This paper studies the privacy risk perception of online medical community users based on deep neural network. Firstly, this paper introduces privacy protection based on deep neural network and users’ privacy risk perception in online medical community. Then, using the fuzzy neural network to deal with highly complex and nonlinear data, we can better obtain the accurate evaluation value, and use the improved gravity search optimization algorithm to optimize the fuzzy neural network evaluation model and improve the convergence puzzle of the model. Finally, using the experimental method of questionnaire survey, and the questionnaire is composed of three parts. The first part investigates the basic personal information of the subjects, including gender, age, educational background, physical condition, physical examination frequency, Internet use experience, long-term residence, etc.; The second part is the measurement items of each variable in the theoretical model, including nine variables: service quality, personalized service, reciprocal norms, result expectation, material reward, perceived risk, trust in doctors, trust in websites, and willingness to disclose health privacy information. The experimental results show that the correlation coefficient between the interaction items of personalized service and reciprocal norms on material reward is positive (β = 0.072, P < 0.01), and the correlation coefficient between sexual service and material reward was positive (β = 0.202, P < 0.01), then reciprocal norms positively regulate the relationship between personalized service and material reward.

Privacy Preserving Heart Rate Estimation from ECG and PPG Signals for Application in Remote Healthcare

“…Yuan et al [33] explored differentially private stochastic gradient descent for X-Ray data, whereas a combination of KL-Divergence and differential privacy was proposed in [34]. Yin et al [35] studied the privacy risk perception of online medical community users based on deep neural network. In addition to such centralizeddata settings, differential privacy has also been employed in the federated learning setting [31,33].…”