TSExplain: Explaining Aggregated Time Series by Surfacing Evolving Contributors

Aggregated time series can be generated effortlessly everywhere, e.g., "total confirmed covid-19 cases since 2019", "S&P500 during the year 2020", and "total liquor sales over time". Understanding "how" and "why" these key performance indicators(KPI) evolve over time is critical to making data-informed decisions. Existing explanation engines focus on explaining the difference between two relations. However, this falls short of explaining KPI's continuous changes over time, as it overlooks explanations in between by only looking at the two endpoints. Motivated by this, we propose TSExplain, a system that explains aggregated time series by surfacing the underlying evolving top contributors. Under the hood, we leverage the existing work on two-relations diff as a building block and formulate a 𝐾-Segmentation problem to segment the time series such that each segment after segmentation shares consistent explanations, i.e., contributors. To quantify consistency in each segment, we propose a novel within-segment variance design based on top explanations; to derive the optimal 𝐾-Segmentation scheme, we develop a dynamic programming algorithm. Experiments on synthetic and real-world datasets show that our explanation-aware segmentation can effectively identify evolving explanations for aggregated time series and outperform explanation-agnostic segmentation. Further, we proposed an optimal selection strategy of 𝐾 and several optimizations to speed up TSExplain for interactive user experience, achieving up to 13× efficiency improvement.