Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Raul Suzuki, Rodrigo Moreira, Pedro Henrique A. Damaso de Melo, Larissa F. Rodrigues Moreira, Flávio de Oliveira Silva

Abstract

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.

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

Abstract

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.

Paper Structure

This paper contains 13 sections, 7 equations, 10 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Proposed Approach
  4. Data and Problem Formulation
  5. Feature engineering (Phase 1)
  6. Per-trace statistics
  7. Path-level temporal context
  8. Rolling and aggregate context features
  9. Base learners and baseline models (Phases 2 and 4)
  10. Stacked ensemble and meta-features (Phase 3)
  11. Training, hyperparameter optimization, and threshold calibration
  12. Results and Discussion
  13. Concluding Remarks

Figures (10)

  • Figure 1: Stacked ensemble architecture (Phases 1--3) and additional baseline models (Phase 4) trained directly on the engineered features for comparison.
  • Figure : (a) CPU utilization vs. training duration.
  • Figure : (a) Training duration versus CPU utilization.
  • Figure : (a) Distribution of F1 scores across ten experimental rounds.
  • Figure : (a) CPU utilization vs. training duration.
  • ...and 5 more figures