From Prototypes to Sparse ECG Explanations: SHAP-driven Counterfactuals for Multivariate Time-series Multi-class Classification
output:af912fac660e7448e78f9e1eThis dossier keeps one entity, its published attributes, every direct edge and the supporting source snapshots under a stable URL.
Identity extends only as far as the published stable identifier or explicit mapping. Similar names are not merged. Missing links mean not observed in the bounded sources—not absent in reality.
identity ≠ name similarity · edge ≠ causalityEvery direct connection and its proof
The relation label and evidence class describe the entire claim. Open the source record before drawing a broader conclusion.
edge:project-output:101120406:af912fac660e7448e78f9e1eNo additional excerpt published.
Open source ↗Fields retained with this identity
Attributes are descriptive source fields. They do not create cross-dataset identity unless an edge states that join explicitly.
- Citation Count
- 0
- Is Open Access
- true
- Open Aire Id
- doi_dedup___::a4ac52d04958c0b2c2ebfd6a7a6be859
- Output Type
- publication
- Persistent Identifiers
- 2510.19514 · 10.1007/s10796-026-10711-9 · 10.48550/arxiv.2510.19514
- Publication Date
- 2026-03-11
Inspect the source snapshots
A source snapshot describes what was available at collection time. Corrections and later publications can change future observations.
Return to the full evidence network
Search every published entity, inspect one-hop neighbourhoods and find the bridges public data still cannot support.