F air -C ount -M in : Frequency Estimation under Equal Group-wise Approximation Factor

Frequency estimation in streaming data often relies on sketches like Count-Min to provide approximate answers with sublinear space. However, Count-Min sketches introduce additive errors that disproportionately impact the unpopular groups, creating fairness concerns. To address these concerns, we introduce Fair-Count-Min, a frequency estimation sketch that guarantees equal expected approximation factors across various groups. We propose a column partitioning approach with group-aware semi-uniform hashing to eliminate collisions between elements from different groups. We provide theoretical guarantees for fairness, analyze its associated cost, and validate our findings through extensive experiments on real-world datasets in comparison with representative state-of-the-art baselines. Our experimental results demonstrate that Fair-Count-Min achieves fairness with generally small additional error while maintaining efficiency comparable to that of the Count-Min sketch.