Efficient sample selection in data stream regression employing evolving generalized fuzzy models

摘要

In this paper, we propose two criteria for efficient sample selection in case of data stream regression problems. The selection becomes apparent whenever the target values, which guide the update of the regressors as well as the implicit model structures, are costly to measure. Reducing the samples used for model updates as much as possible while keeping the predictive accuracy of the models on a high level is thus a central challenge, especially in non-stationary environments where (permanent) system changes or expansion can be expected. Our selection criteria rely on two aspects: 1.) the extrapolation degree of the model combined with its non-linearity degree, 2.) the uncertainty in model outputs which can be measured in terms of confidence intervals reflected by so-called adaptive error bars, which are updated over time synchronously to the model. The selection criteria are developed in combination with evolving generalized Takagi-Sugeno (TS) fuzzy models (containing rules in arbitrarily rotated position), which could be shown to outperform conventional evolving TS models (containing axis-parallel rules) and other stream regression techniques in previous publications. The results based on two high-dimensional real-world streaming problems show that a decrease of the number of model updates by about 80-85% (as only 15-20% of samples are selected) can still achieve similar accumulated model errors over time to the case when performing a full update on all samples. This may yield a significant reduction of computational demands and of costs whenever targets are costly to measure.