Matching Detections to Events in Time Series
SoftED metrics introduce a soft evaluation of event detection methods in time series, incorporating fuzzy logic concepts to provide temporal tolerance in detections. However, these metrics face challenges associating detections with events, especially in cases with multiple associations between detections and events. In this work, we propose structuring this association problem within the graph theory paradigm, approaching it as a bipartite graph matching problem. For this, the Hungarian algorithm is employed to solve the association problem. The results demonstrate the effectiveness of the proposed approach, highlighting the impact of improvements in the associations between detections and events.
Authors
- Marco Antonio Moura Reis (ORCID: https://orcid.org/0000-0003-4064-0240)
- Rebecca Salles (ORCID: https://orcid.org/0000-0002-1001-3839)
- Eduardo Ogasawara (ORCID: https://orcid.org/0000-0002-0466-0626)
- Rafaelli Coutinho (ORCID: https://orcid.org/0000-0002-1735-1718)
- Geraldo Xexéo (ORCID: https://orcid.org/0000-0003-3975-9076)
Institutions
- Universidade Federal do Rio de Janeiro (BR)
- Institut national de recherche en sciences et technologies du numérique (FR)
- Federal Center for Technological Education Celso Suckow da Fonseca (BR)
Publication Details
- Published
- 2024-10-14
- DOI
- https://doi.org/10.5753/sbbd.2024.243275
- Primary Topic
- Anomaly Detection Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Coordenação de Aperfeiçoamento de Pessoal de Nível Superior
- Conselho Nacional de Desenvolvimento Científico e Tecnológico
- Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro