Abstrakti

Smoothing Outlier Scores is All You Need to Improve Outlier Detectors (Extended Abstract)




TekijätYang, Jiawei; Rahardja, Susanto; Fränti, Pasi

ToimittajaN/A

Konferenssin vakiintunut nimiInternational Conference on Data Engineering

KustantajaIEEE

Julkaisuvuosi2025

Kokoomateoksen nimi2025 IEEE 41st International Conference on Data Engineering (ICDE)

Aloitussivu4736

Lopetussivu4737

DOIhttps://doi.org/10.1109/ICDE65448.2025.00397

Verkko-osoitehttps://doi.org/10.1109/icde65448.2025.00397


Tiivistelmä

Existing outlier detectors calculate outlier scores for data objects independently, ignoring the consistency between score similarity and object similarity. As a result, these detectors may produce inconsistent scores for similar objects, leading the scores of some normal objects to exceed some of outlier objects, increasing the possibility of misclassification. To address this issue, we first assume that similar objects should have similar scores. Then, based on this assumption, we propose neighborhood averaging, an outlier score post-processing technique to improve any single outlier detector, which is the first of its kind.


Julkaisussa olevat rahoitustiedot
This work has been co-funded by the European Union's Horizon Europe Framework programme for research and in-novation 2021–2027 under the Marie Skłodowska-Curie grant agreement n° 101126611.


Last updated on 2025-06-10 at 09:02