Outlier Ranking in Large-Scale Public Health Streams
This work addresses a practical bottleneck for disease control experts in public health by enabling faster outlier investigation, though it is incremental as it builds on existing univariate methods.
The paper tackled the problem of experts being overwhelmed by thousands of tied outliers in large-scale public health data streams by proposing a new algorithm to rank these outliers, resulting in experts identifying important outliers 9.1 times faster than their prior baseline.
Disease control experts inspect public health data streams daily for outliers worth investigating, like those corresponding to data quality issues or disease outbreaks. However, they can only examine a few of the thousands of maximally-tied outliers returned by univariate outlier detection methods applied to large-scale public health data streams. To help experts distinguish the most important outliers from these thousands of tied outliers, we propose a new task for algorithms to rank the outputs of any univariate method applied to each of many streams. Our novel algorithm for this task, which leverages hierarchical networks and extreme value analysis, performed the best across traditional outlier detection metrics in a human-expert evaluation using public health data streams. Most importantly, experts have used our open-source Python implementation since April 2023 and report identifying outliers worth investigating 9.1x faster than their prior baseline. Other organizations can readily adapt this implementation to create rankings from the outputs of their tailored univariate methods across large-scale streams.