2026-04-20T02:02:11Zhttps://recercat.cat/oai/request

oai:recercat.cat:2445/2180422025-12-05T09:55:37Zcom_2072_1057col_2072_478917col_2072_478920

urn:hdl:2445/218042 Importance attribution in neural networks by means of persistence landscapes of time series Ferrà Marcús, Aina Casacuberta, Carles Pujol Vila, Oriol Xarxes neuronals (Informàtica) Anàlisi de sèries temporals Homologia Neural networks (Computer science) Time-series analysis Homology This article describes a method to analyze time series with a neural network using a matrix of area-normalized persistence landscapes obtained with topological data analysis. The network’s architecture includes a gating layer that is able to identify the most relevant landscape levels for a classification task, thus working as an importance attribution system. Next, a matching is performed between the selected landscape levels and the corresponding critical points of the original time series. This matching enables reconstruction of a simplified shape of the time series that gives insight into the grounds of the classification decision. As a use case, this technique is tested in the article with input data from a dataset of electrocardiographic signals. The classification accuracy obtained using only a selection of landscape levels from data was 94.00% averaged after five runs of a neural network, while the original signals achieved 98.41% and landscape-reduced signals yielded 97.04%. 2025-01-28T09:01:55Z 2025-01-28T09:01:55Z 2023-07-19 2025-01-28T09:01:56Z info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion Reproducció del document publicat a: https://doi.org/10.1007/s00521-023-08731-6 Neural Computing & Applications, 2023, vol. 35, p. 20143-20156 https://doi.org/10.1007/s00521-023-08731-6 cc by (c) Aina Ferrà Marcús et al., 2023 http://creativecommons.org/licenses/by/3.0/es/ info:eu-repo/semantics/openAccess Springer Verlag Articles publicats en revistes (Matemàtiques i Informàtica)