论文信息 - Relevance learning for time series inspection

Relevance learning for time series inspection

By means of local neighborhood regression and time windows, the generative topographic mapping (GTM) allows to predict and visually inspect time series data. GTM itself, however, is fully unsupervised. In this contribution, we propose an extension of relevance learning to time series regression with GTM. This way, the metric automatically adapts according to the relevant time lags resulting in a sparser representation, improved accuracy, and smoother visualization of the data.

Amaury Lendasse | Barbara Hammer | Andrej Gisbrecht | Dusan Sovilj

[1] Guilherme De A. Barreto,et al. Time Series Prediction with the Self-Organizing Map: A Review , 2007, Perspectives of Neural-Symbolic Integration.

[2] Christopher M. Bishop,et al. GTM: The Generative Topographic Mapping , 1998, Neural Computation.

[3] Holger Kantz,et al. Nonlinear Time Series Analysis: Preface to the second edition , 2003 .

[4] Barbara Hammer,et al. Generalized Relevance LVQ for Time Series , 2001, ICANN.

[5] Barbara Hammer,et al. Relevance learning in generative topographic mapping , 2011, Neurocomputing.

[6] Thomas Villmann,et al. Generalized relevance learning vector quantization , 2002, Neural Networks.

[7] Michel Verleysen,et al. Forecasting the CATS benchmark with the Double Vector Quantization method , 2007, Neurocomputing.

[8] Michel Verleysen,et al. High-dimensional delay selection for regression models with mutual information and distance-to-diagonal criteria , 2007, Neurocomputing.