Nukhba Nukhba - ANOMALY DETECTION IN NETWORK MONITORING - A comparison of performance analysis of generic vs cloud-based Machine learning Platform

Aug 3, 2026

Upload by Patrick

Anomaly detection techniques for telecommunications enable automated network supervision and improved reliability with reduced operational cost. This Master’s thesis compares a generic machine learning platform, scikit-learn, with Azure Machine Learning Studio using anomaly detection as the central use case. Both platforms apply multiple machine learning methods to two anomalously labeled network datasets and are evaluated across data processing, model creation and prediction, cost, and usability. Findings indicate cloud platforms offer stronger usability through low/no-code workflows, but reduced control and limited ready-to-use and unsupervised support; cloud deployment remains production-ready with minimal effort.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages55

Reading Time2 h 19 min

AccessFree to download

Introduction

Structure of the Thesis

Background

Anomaly Detection

Measures for Anomaly Detection

Machine learning Methods

Machine Learning as a Service

Research Objective

Related Work

Density based and Clustering techniques

Classification Based techniques

Supervised anomaly detection techniques

Methodology

Anomaly detection workflow

Data

Dimensionality reduction

Models

Unsupervised Learning K-means clustering

Nukhba Nukhba ANOMALY DETECTION IN NETWORK MONITORING a comparison of performance analysis of generic vs cloud-based Machine learning Platform Master of Science Thesis Faculty of Information Technology and Communication Sciences Examiners: Ph.D. Tapio Elomaa, D.Sc. (Tech) Adrian Burian May 2023 ABSTRACT Nukhba Nukhba: Anomaly Detection in Network Monitoring, a comparison of performance analysis of generic vs cloud-based Machine learning Platform M. Sc. Thesis Tampere University Master’s Degree Programme in May 2023 Applications of anomaly detection in telecommunication industry are widely in demand for the purpose of automating network supervision and to provide better network reliability at reduced cost. To meet this increasing demand, various machine learning platforms have emerged that empower businesses to leverage ML capabilities at a better cost. This study compares a generic machine learning platform scikit-learn with Azure Machine learning studio, while taking anomaly detection as a centre use case. The focus of this thesis is to understand the value offered by the Cloud giants for the application of anomaly detection and possibly suggest a better platform for the development of machine learning application, deployment, and maintenance. This thesis studies a generic and cloud-based machine learning platform by employing various machine learning methods for anomaly detection. The machine learning methods from a generic platform and a cloud-based platform are applied onto two anomalous labelled network datasets. These platforms are analysed with respect to data processing, model creation and prediction, cost, and usability. As a result, this work concludes the comparison with various aspects of model creation, visualization of the results, model performance, cost, and ease of use. Results show that cloud-based machine learning platforms do provide an edge with respect to usability by providing low to no code option. However, low code also implies less control over the model, and it also comes with limited ready to use algorithms. And even less built-in support for unsupervised algorithms in Azure designer. The whole process of model creation, data processing and deployment was effortless, thus, it is easy to convert to production ready. In addition, cloud-based platforms eliminate the need of buying high end machine for storage computation. However, the studio charges money per experimentation hour and storage. Both platforms are well suited for the purpose of anomaly detection with each having some advantage over the other. Keywords and terms: Anomaly detection, mobile networks, machine learning. The originality of this thesis has been checked using the Turnitin OriginalityCheck service. PREFACE This degree project is carried out at Nokia Corporation, a Finnish multinational, telecommunications, information technology and consumer electronics corporation with global existence. Nokia Networks is one of the biggest subsidiaries of Nokia Corp and is a leading vendor in data networking, IP infrastructure, telecommunication equipment, software and related services used by communication service providers. The objective of this study is to analyse the application of anomaly detection system for network monitoring, as part of a project taking place inside Nokia Networks. CONTENTS 1. INTRODUCTION.......................................................................................................................9 1.1 Structure of the Thesis..................................................................................................... 10 2. BACKGROUND ......................................................................................................................... 11 2.1 Anomaly Detection ............................................................................................................ 11 2.2 Measures for Anomaly Detection ....................................

FAQs

What platforms are compared in this thesis for anomaly detection?

The study compares a generic machine learning platform (scikit-learn) with Azure Machine Learning Studio, using anomaly detection as the main use case.

How are the platforms evaluated in the thesis?

Evaluation covers data processing, model creation and prediction, cost, and usability, using two anomalous labeled network datasets.

What are the key conclusions about cloud-based platforms versus generic tools?

Cloud-based platforms provide an advantage in usability via low/no-code options and reduce hardware needs, but they also limit model control and offer restricted ready-to-use and unsupervised algorithm support.

Aug 3, 2026

Upload by Patrick