{"id":116952,"date":"2026-09-11T17:02:51","date_gmt":"2026-09-11T15:02:51","guid":{"rendered":"https:\/\/nilu.gnist.dev\/publikasjoner\/supervised-anomaly-detection-in-univariate-time-series-using-1d-convolutional-siamese-networks\/"},"modified":"2026-09-11T17:02:51","modified_gmt":"2026-09-11T15:02:51","slug":"supervised-anomaly-detection-in-univariate-time-series-using-1d-convolutional-siamese-networks","status":"publish","type":"nva_publication","link":"https:\/\/nilu.gnist.dev\/en\/publications\/supervised-anomaly-detection-in-univariate-time-series-using-1d-convolutional-siamese-networks\/","title":{"rendered":"Supervised Anomaly Detection in Univariate Time-Series Using 1D Convolutional Siamese Networks"},"content":{"rendered":"<p class=\"wp-block-paragraph\">In time-series data analysis, identifying anomalies is crucial for maintaining data integrity and ensuring accurate analyses and decision-making. Anomalies can compromise data quality and operational efficiency. The complexity of time-series data, with its temporal dependencies and potential non-stationarity, makes anomaly detection challenging but essential. Our research introduces ADSiamNet, a 1D Convolutional Neural Network-based Siamese network model for anomaly detection and rectification. ADSiamNet effectively identifies localized patterns in time-series data and smooths detected anomalies using a quantile-based technique. In tests with physical activity data from Actigraph watches and MOX2-5 sensors, ADSiamNet achieved accuracies of 98.65% and 85.0%, respectively, outperforming other supervised anomaly detection methods. The model uses a contrastive loss function to compare input sequences and adjusts network weights iteratively during training to recognize intricate patterns. Additionally, we evaluated various univariate time-series forecasting algorithms on datasets with and without anomalies. Results show that anomaly-smoothed data reduces forecasting errors, highlighting our approach\u2019s effectiveness in enhancing time-series data analysis\u2019s integrity and reliability. Future research will focus on multivariate time-series datasets.<\/p>\n","protected":false},"template":"","meta":{"_acf_changed":false,"_searchwp_excluded":"","_id":"","_status":"PUBLISHED","_sync_date":"2026-09-11 14:14:22","footnotes":""},"nva_tax_category":[1099],"class_list":["post-116952","nva_publication","type-nva_publication","status-publish","hentry","nva_tax_category-scientific-journal-publication"],"acf":[],"_links":{"self":[{"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/nva_publication\/116952","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/nva_publication"}],"about":[{"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/types\/nva_publication"}],"version-history":[{"count":0,"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/nva_publication\/116952\/revisions"}],"wp:attachment":[{"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/media?parent=116952"}],"wp:term":[{"taxonomy":"nva_tax_category","embeddable":true,"href":"https:\/\/nilu.gnist.dev\/en\/wp-json\/wp\/v2\/nva_tax_category?post=116952"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}