Data Augmentation (DA) -- enriching training data by adding synthetic samples
-- is a technique widely adopted in Computer Vision (CV) and Natural Language
Processing (NLP) tasks to improve models performance. Yet, DA has struggled to
gain traction in networking contexts, particularly in Traffic Classification
(TC) tasks. In this work, we fulfill this gap by benchmarking 18 augmentation
functions applied to 3 TC datasets using packet time series as input
representation and considering a variety of training conditions. Our results
show that (i) DA can reap benefits previously unexplored with (ii)
augmentations acting on time series sequence order and masking being a better
suit for TC and (iii) simple latent space analysis can provide hints about why
augmentations have positive or negative effects.

数据增强（DA）在计算机视觉（CV）和自然语言处理（NLP）领域被广泛采用以提高模型性能，然而在网络环境中，尤其是在流量分类（TC）任务中，数据增强一直存在问题。本文通过对 3 个 TC 数据集使用数据包时间序列作为输入表示，并考虑多种训练条件，基准测试了 18 种增强函数，结果显示（i）DA 可以获得以前未探索的优势，（ii）基于时间序列顺序和掩码的增强函数更适用于 TC，（iii）简单的潜在空间分析可以提供有关增强效果的线索。