EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Majority Voting Versus Learned-Threshold Score Aggregation for Verifying IoT Security Events Under Non-IID Heterogeneity

Ivan Ortiz-Garces, Fernando Molina-Granja iD, Rommel Gutierrez iD, Fernando Vela-Espin

Sensors · 04.11.2026 · 26 (21) · 6969

DOI: 10.3390/s26216969 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 04.11.2026. Verfasst von Ivan Ortiz-Garces und 3 weiteren Autor:innen. Beteiligte Einrichtung: Escuela de Ingeniería en Ciberseguridad, Universidad de Las Américas, Quito 170124, Ecuador. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Intrusion detectors for Internet of Things (IoT) systems are usually trained per device or site, so a verification layer must combine alerts from detectors built on different populations and base rates. This study compares two ways of turning such alerts into a decision: majority voting with a fixed threshold, and score aggregation with a threshold learned on labeled calibration data. Seven detectors trained separately on TON_IoT device populations voted on 56,000 shared events, and each rule was compared with the detector local to the event through an amplification factor A. Fixed-threshold majority voting amplified error in the point estimate (A=1.12 for 1−MCC), whereas learned-threshold aggregation reduced it (A≈0.62); device-level intervals, however, included neutrality. An equal-weight score with a learned threshold matched reliability-, margin-, and hybrid-weighted rules, so the improvement came from threshold placement rather than from weighting. In this setting, about 500 labeled target events sufficed to place the threshold, whereas source-domain reliability estimates were biased by up to 0.22 although their 95% sampling half-width was at most 0.011. Gains concentrated in devices with weak local detectors and shrank under natural prevalence. Under degraded validators, fixed-threshold voting was the most fragile rule. A synthetic vote layer overstated amplification, and a federated detector showed a shared threshold collapsing under domain shift.

Schlagwörter

Stichwörter der Autor:innen

IoT intrusion detectionNon-IID heterogeneitycalibrationcompromised validatorsdetector aggregationmajority votingthreshold learning

Autor:innen und Institutionen

  • Escuela de Ingeniería en Ciberseguridad, Universidad de Las Américas, Quito 170124, Ecuador.
  • Facultad de Ingeniería, Universidad Nacional de Chimborazo, Riobamba 060108, Ecuador.

ORCID-Profile: Fernando Molina-Granja, Rommel Gutierrez

Zitieren

APA

Ortiz-Garces, I., Molina-Granja, F., Gutierrez, R., & Vela-Espin, F. (2026). Majority Voting Versus Learned-Threshold Score Aggregation for Verifying IoT Security Events Under Non-IID Heterogeneity. Sensors, 26(21), 6969. https://doi.org/10.3390/s26216969

Vancouver

Ortiz-Garces I, Molina-Granja F, Gutierrez R, Vela-Espin F. Majority Voting Versus Learned-Threshold Score Aggregation for Verifying IoT Security Events Under Non-IID Heterogeneity. Sensors. 2026;26(21):6969. doi:10.3390/s26216969

BibTeX
@article{OrtizGarces2026,
  author = {Ortiz-Garces, Ivan and Molina-Granja, Fernando and Gutierrez, Rommel and Vela-Espin, Fernando},
  title = {Majority Voting Versus Learned-Threshold Score Aggregation for Verifying IoT Security Events Under Non-IID Heterogeneity},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {6969},
  doi = {10.3390/s26216969}
}
Kostenlos starten →