# Drift-Based Dataset Stability Benchmark

arXiv: https://arxiv.org/abs/2512.23762 · HTML: https://mlsanity.com/paper/2512.23762/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 37 references that are in the map.
- First posted on arXiv 28 Dec 2025.

## Builds on (most cited first, 9 of 9)

- [WILDS: A Benchmark of in-the-Wild Distribution Shifts](https://mlsanity.com/paper/2012.07421/) (2020)
- [Learning under Concept Drift: A Review](https://mlsanity.com/paper/2004.05785/) (2020)
- [Fine-grained TLS services classification with reject option](https://mlsanity.com/paper/2202.11984/) (2022)
- [Concept Drift Detection from Multi-Class Imbalanced Data Streams](https://mlsanity.com/paper/2104.10228/) (2021)
- [Datasets are not Enough: Challenges in Labeling Network Traffic](https://mlsanity.com/paper/2110.05977/) (2021)
- [LARA: A Light and Anti-overfitting Retraining Approach for Unsupervised Time Series Anomaly Detection](https://mlsanity.com/paper/2310.05668/) (2023)
- [NetTiSA: Extended IP Flow with Time-series Features for Universal Bandwidth-constrained High-speed Network Traffic Classification](https://mlsanity.com/paper/2310.05530/) (2023)
- [Padding Ain't Enough: Assessing the Privacy Guarantees of Encrypted DNS](https://mlsanity.com/paper/1907.01317/) (2019)
- [DataZoo: Streamlining Traffic Classification Experiments](https://mlsanity.com/paper/2310.19568/) (2023)

## Abstract (arXiv)

Machine learning (ML) represents an efficient and popular approach for network traffic classification. However, network traffic classification is a challenging domain, and trained models may degrade soon after deployment due to the obsolete datasets and quick evolution of computer networks as new or updated protocols appear. Moreover, significant change in the behavior of a traffic type (and, therefore, the underlying features representing the traffic) can produce a large and sudden performance drop of the deployed model, known as a data or concept drift. In most cases, complete retraining is performed, often without further investigation of root causes, as good dataset quality is assumed. However, this is not always the case and further investigation must be performed. This paper proposes a novel methodology to evaluate the stability of datasets and a benchmark workflow that can be used to compare datasets. The proposed framework is based on a concept drift detection method that also uses ML feature weights to boost the detection performance. The benefits of this work are demonstrated on CESNET-TLS-Year22 dataset. We provide the initial dataset stability benchmark that is used to describe dataset stability and weak points to identify the next steps for optimization. Lastly, using the proposed benchmarking methodology, we show the optimization impact on the created dataset variants.