TRACE: Traceroute-based Internet Route change Analysis with Ensemble Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Suzuki, Raul, Moreira, Rodrigo, de Melo, Pedro Henrique A. Damaso, Moreira, Larissa F. Rodrigues, Silva, Flávio de Oliveira
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911564248907776
author Suzuki, Raul
Moreira, Rodrigo
de Melo, Pedro Henrique A. Damaso
Moreira, Larissa F. Rodrigues
Silva, Flávio de Oliveira
author_facet Suzuki, Raul
Moreira, Rodrigo
de Melo, Pedro Henrique A. Damaso
Moreira, Larissa F. Rodrigues
Silva, Flávio de Oliveira
contents Detecting Internet routing instability is a critical yet challenging task, particularly when relying solely on endpoint active measurements. This study introduces TRACE, a MachineLearning (ML)pipeline designed to identify route changes using only traceroute latency data, thereby ensuring independence from control plane information. We propose a robust feature engineering strategy that captures temporal dynamics using rolling statistics and aggregated context patterns. The architecture leverages a stacked ensemble of Gradient Boosted Decision Trees refined by a hyperparameter-optimized meta-learner. By strictly calibrating decision thresholds to address the inherent class imbalance of rare routing events, TRACE achieves a superior F1-score performance, significantly outperforming traditional baseline models and demonstrating strong effective ness in detecting routing changes on the Internet.
format Preprint
id arxiv_https___arxiv_org_abs_2604_02361
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle TRACE: Traceroute-based Internet Route change Analysis with Ensemble Learning
Suzuki, Raul
Moreira, Rodrigo
de Melo, Pedro Henrique A. Damaso
Moreira, Larissa F. Rodrigues
Silva, Flávio de Oliveira
Networking and Internet Architecture
Artificial Intelligence
Machine Learning
Detecting Internet routing instability is a critical yet challenging task, particularly when relying solely on endpoint active measurements. This study introduces TRACE, a MachineLearning (ML)pipeline designed to identify route changes using only traceroute latency data, thereby ensuring independence from control plane information. We propose a robust feature engineering strategy that captures temporal dynamics using rolling statistics and aggregated context patterns. The architecture leverages a stacked ensemble of Gradient Boosted Decision Trees refined by a hyperparameter-optimized meta-learner. By strictly calibrating decision thresholds to address the inherent class imbalance of rare routing events, TRACE achieves a superior F1-score performance, significantly outperforming traditional baseline models and demonstrating strong effective ness in detecting routing changes on the Internet.
title TRACE: Traceroute-based Internet Route change Analysis with Ensemble Learning
topic Networking and Internet Architecture
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2604.02361