Learning from Time-Changing Data with Adaptive Windowing

Abstract

We present a new approach for dealing with distribution change and concept drift when learning from data sequences that may vary with time. We use sliding windows whose size, instead of being fixed a priori, is recomputed online according to the rate of change observed from the data in the window itself. This delivers the user or programmer from having to guess a time-scale for change. Contrary to many related works, we provide rigorous guarantees of performance, as bounds on the rates of false positives and false negatives.Using ideas from data stream algorithmics, we develop a time-and memory-efficient version of this algorithm, called ADWIN2. We show how to combine ADWIN2 with the Naïve Bayes (NB) predictor, in two ways: one, using it to monitor the error rate of the current model and declare when revision is necessary and, two, putting it inside the NB predictor to maintain up-to-date estimations of conditional probabilities in the data. We test our approach using synthetic and real data streams and compare them to both fixed-size and variable-size window strategies with good results.

An Efficient Concept Drift Detection Method for Streaming Data under Limited Labeling

“…In [14] learning from concept-drifting data streams is performed based on a sliding window of an adaptive size. The window size increases when no change is apparent and it shrinks when data changes.…”

“…By assigning weights to ensemble members depending on the estimated error rate, concept drift can be dealt with [15]. In [16], two ensemble methods, bagging by ADWIN (ADaptive WINdowing) [14] and bagging by adaptive-size Hoeffding trees, are proposed.…”

Online active supervision of an evolving classifier for customized-gesture-command learning

“…A method to adapt to concept drifts in the data stream is to use a sliding window that only contains most recent data samples [13] [14]. The use of a sliding window also has the advantage of limiting both the computational cost and the memory requirement.…”

On the Online Classification of Data Streams Using Weak Estimators

“…However, in case of an increasing error rate that reaches the second threshold, a concept change is declared and the learning model is retrained from the buffered data that appeared after t w . [2,3] proposed an adaptive sliding window scheme named ADWIN for change detection and for estimating statistics from the data stream. It was shown that the ADWIN algorithm outperforms the SPC approach and that it has the ability to provide rigorous guarantees on false positive and false negative rates.…”