Time Series Privacy-Preserving: New Approaches via Complex Networks

TSP2Net is an exploratory project dedicated to developing new ways to protect privacy in time series without compromising data utility. The proposal is based on the use of complex networks as an alternative representation of these data. By transforming time series into networks, the project aims to create a safer way of sharing data, where the original structure of the data - that is, sensitive information - is less exposed. TSP2Net also investigates methods capable of performing the reverse process: generating synthetic time series from the produced networks. These series preserve the essential properties of the original data, allowing effective exploration and analysis while minimizing the risks of identification or disclosure of confidential information. Overall, the project explores new non-parametric approaches that combine network science and time series analysis, contributing to a more robust balance between privacy and utility in the sharing of temporal data.