Dataset Quality Assessment in Autonomous Networks with Permutation Testing

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F22%3A00358730" target="_blank" >RIV/68407700:21240/22:00358730 - isvavai.cz</a>

Result on the web

<a href="https://pesw.fit.cvut.cz/2022/PESW_2022.pdf" target="_blank" >https://pesw.fit.cvut.cz/2022/PESW_2022.pdf</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Dataset Quality Assessment in Autonomous Networks with Permutation Testing

Original language description

The development of autonomous or self-driving networks is one of the main challenges faced by the telecommunication industry. Future networks are expected to realise a number of tasks, including network optimization and failure recovery, with minimal human supervision. In this context, the network community (manufacturers, operators, researchers, etc.) is looking at Machine Learning (ML) methods with high expectations. However, ML models can only be as good as the datasets they are trained on, which means that autonomous networks also require a sound autonomous procedure to assess, and if possible improve, dataset quality. Although the application of ML techniques in communication networks is ample in the literature, analyzing the quality of the network datasets seems an ignored problem. This paper presents work in progress on the application of permutation testing to assess the quality of network datasets. We illustrate our approach on a number of simple synthetic datasets with pre-established quality and then we demonstrate its application in a publicly available network datasets.

Czech name

—

Czech description

—

Type

O - Miscellaneous

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/VJ02010024" target="_blank" >VJ02010024: Flow-based Encrypted Traffic Analysis</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2022

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů