PPDCA: Privacy-preserving crowdsensing data collection and analysis with randomized response

Randomized response mechanisms for guaranteeing crowdsensing data privacy have attracted scholarly attention; aggregators can ensure privacy by collecting only randomized data and individuals have plausible deniability regarding their responses. The analysts employed by organizations can still make predictions and conduct analyses using the randomized data. Existing randomized response-based data collection solutions have severely restricted functionality and usability, resulting in impractical and inefficient systems. Hence, we propose a randomized response-based privacy-preserving crowdsensing data collection and analysis (PPDCA) method, in which a complementary randomized response (C-RR) approach is designed to guarantee data privacy and to preserve features for data analysis. Moreover, we transform encoded data into binary vectors and generate a learning network using a deep learning framework. Through C-RR and our learning model, PPDCA can perform exceptionally in terms of high-utility analysis for the collected client-side strings, compared with state-of-the-art methods.