In the present work, a technique for reducing the dimensionality of tensor data, called Low-Correlation Multilinear Dimensionality Reduction (LC-MDR), is proposed. The method optimizes a cost function that takes into account the data correlation, generating variables with low correlation. The LC-MDR fits the input data correlation into a new tensor decomposition denoted by Even-Order Nested PARAFAC Decomposition (EONPD), a proposed extension of the Nested PARAFAC Decomposition (NPD) for higher-order tensors. In addition, a generalization of the EONPD, denoted by Higher-Order Nested PARAFAC Decomposition (HONPD), is also presented. Contrarily to existing approaches that use orthogonal transformation matrices, the LC-MDR does not impose this constraint on the transformation matrices in order to minimize the output correlation. The proposed technique was evaluated in a classification system of volcano-seismic events using data obtained from the Ubinas volcano in 2009 using a full tensorial classification framework. The results showed significant gains for the LC-MDR when compared with concurrent techniques in terms of accuracy, data correlation and processing time.

On 14 August 2021, the moment magnitude (M-w) 7.2 Nippes earthquake in Haiti occurred within the same fault zone as its devastating 2010 M-w 7.0 predecessor, but struck the country when field access was limited by insecurity and conventional seismometers from the national network were inoperative. A network of citizen seismometers installed in 2019 provided near-field data critical to rapidly understand the mechanism of the mainshock and monitor its aftershock sequence. Their real-time data defined two aftershock clusters that coincide with two areas of coseismic slip derived from inversions of conventional seismological and geodetic data. Machine learning applied to data from the citizen seismometer closest to the mainshock allows us to forecast aftershocks as accurately as with the network-derived catalog. This shows the utility of citizen science contributing to our understanding of a major earthquake.

This article proposes a supervised tensor-based learning framework for classifying volcano-seismic events from signals recorded at the Ubinas volcano, in Peru, during a period of great activity in 2009. The proposed method is fully tensorial, as it integrates the three main steps of the automatic classification system (feature extraction, dimensionality reduction, and classifier) in a general multidimensional framework for tensor data, joining tensor learning techniques such as the multilinear principal component analysis (MPCA) and the support tensor machine (STM). By exploiting the use of multiple multichannel triaxial sensors, operating simultaneously in two seismic stations, the tensor patterns are constructed as stations × channels × features . The multidimensional structure of the data is then preserved, avoiding the tensor vectorization that often leads to a feature vector with a large dimension, which increases the number of parameters and may cause the “curse of dimensionality.”Moreover, the array vectorization breaks down the multidimensional structure of the data, which usually leads to performance degradation. The results showed a good performance of the proposed multilinear classification system, significantly outperforming its vectorial counterparts. The best result was obtained with the STuM classifier along with the MPCA.

The evaluation and prediction of volcanoes activities and associated risks is still a timely and open issue. The amount of volcano-seismic data acquired by recent monitoring stations is huge (e.g., several years of continuous recordings), thereby making machine learning absolutely necessary for their automatic analysis. The transient nature of the volcano-seismic signatures of interest further enforces the need of automatic detection and classification of such events. In this paper, we present a novel architecture for automatic classification of volcano-seismic events based on a comprehensive signal representation with a large feature set. To the best of our knowledge this is one of the first attempts to automatize the classification task of these signals. The proposed approach relies on supervised machine learning techniques to build a prediction model.