Toward the better modeling and visualization of uncertainty for streaming data

Streaming data can be found in many different scenarios, in which data are generated and arriving continuously. Sampling approaches have been proven as an effective means to cope with the sheer volume of the streaming data. However, sampling methods also introduce uncertainty, which can affect the reliability of subsequent analysis and visualization. In this paper, we propose a novel model called PDm and visualization named uncertainty tree to present uncertainty that arises from sampling streaming data. PDm is first introduced to characterize uncertainty of streaming data, and an optimization method is then proposed to minimize uncertainty. Uncertainty tree is further developed to enhance data understanding by visualizing uncertainty and revealing temporal patterns of streaming data. Lastly, a quantitative evaluation and real-world examples have been conducted to demonstrate the effectiveness and efficacy of the proposed techniques.