Querying Temporal Drifts at Multiple Granularities (Technical Report)

TL;DR

This paper introduces a query-based method for detecting data stream drifts at multiple time granularities using a drift index, enabling flexible and efficient drift queries with demonstrated performance on real and synthetic datasets.

Contribution

It proposes a novel query-based approach with a drift index for multi-granularity drift detection, including formalization, algorithms, and performance evaluation.

Findings

Effective detection of drifts at multiple granularities

Algorithms perform well on real-world and synthetic datasets

Flexible drift query evaluation strategies

Abstract

There exists a large body of work on online drift detection with the goal of dynamically finding and maintaining changes in data streams. In this paper, we adopt a query-based approach to drift detection. Our approach relies on {\em a drift index}, a structure that captures drift at different time granularities and enables flexible {\em drift queries}. We formalize different drift queries that represent real-world scenarios and develop query evaluation algorithms that use different materializations of the drift index as well as strategies for online index maintenance. We describe a thorough study of the performance of our algorithms on real-world and synthetic datasets with varying change rates.