ADVERTISEMENT CLICK FRAUD DETECTION AND PREVENTION - A MACHINE LEARNING APPROACH - Project Work

Aug 2, 2026

Upload by Miles

Click fraud undermines digital advertising by generating substantial financial losses and weakening advertiser trust. This project work investigates how machine learning methods can detect malicious click behavior in Google Ads. Five algorithms—support vector machines, random forest, k-nearest neighbors, gradient tree boosting, and XGBoost—are compared within the CRISP-DM workflow. Results show tree-based models, especially GTB and XGBoost, achieve the strongest accuracy, recall, and AUC. The study links fraudulent patterns to primary click actions and click frequency per IP address and user ID, supporting more effective prevention for marketing agencies.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages47

Reading Time1 h 58 min

AccessFree to download

Introduction

Literature Review

Click Fraud In Online Advertising

Machine Learning Algorithms For Fraud Detection

Conceptual Model

Methodology

Business Understanding

Data Understanding

Data Preparation

Data Splitting

Selection Of Models

Results

Support Vector Machine (Svm)

Random Forest (Rf)

Model K-Nearest Neighbors (Knn)

Model Gradient Tree Boosting (Gtb)

Model Xgboost

Discussion

Theoretical Implications

Practical Implications

Limitations And Future Research

MDDM Master’s Degree Program in Data-Driven Marketing ADVERTISEMENT CLICK FRAUD DETECTION AND PREVENTION A machine learning approach Camilla Alves do Espírito Santo Project Work presented as a partial requirement for obtaining a Master’s Degree in Data-Driven Marketing NOVA Information Management School Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa NOVA Information Management School Instituto Superior de Estatística e Gestão de Informação Universidade Nova de Lisboa ADVERTISEMENT CLICK FRAUD DETECTION AND PREVENTION A MACHINE LEARNING APPROACH By Camilla Alves do Espírito Santo Project Work presented as partial requirement for obtaining the Master’s degree in DataDriven Marketing, with a specialization in Data Science for Marketing. Supervised by Bruno Jardim, PhD, NOVA Information Management School STATEMENT OF INTEGRITY I hereby declare having conducted this academic work with integrity. I confirm that I have not used plagiarism, any form of undue use of information or falsification of results along the process leading to its elaboration. I further declare that I have fully acknowledged the Rules of Conduct and Code of Honor from the NOVA Information Management School. Camilla Alves do Espírito Santo Lisboa, 15/07/2024 ABSTRACT Click fraud poses a significant challenge to digital advertising, causing substantial financial losses and undermining advertiser trust. The study explores the potential of machine learning approaches for detecting such malicious conduct in Google Ads. We use five algorithms for modelling and comparison, including support vector machines, random forest, k-nearest neighbours, gradient tree boosting, and XGBoost. These are all part of the CRISP-DM methodology, which gives you a structured way to do machine learning projects. These models were chosen for their proven efficacy in fraud detection. Our analysis revealed that tree-based models, particularly GTB and XGBoost, outperformed others in accuracy, recall, and AUC, making them highly effective in identifying fraudulent clicks. The study confirms that machine learning algorithms can accurately classify and detect fraudulent activities, enhancing the understanding of fraud characteristics using pre-classified data. Additionally, we identified key patterns and characteristics associated with non-genuine clicks, such as primary click actions and click frequency per IP address and user ID. This research bridges the gap between academic theory and practical application, providing actionable insights for marketing agencies to combat click fraud effectively. A collaboration with a marketing agency for this study ensures that the outcomes are directly beneficial, enhancing the overall integrity and performance of digital advertising efforts. KEYWORDS Click fraud; machine learning; advertising; detection; ads Sustainable Development Goals (SDG): Table Of Contents 1. Introduction .................................................................................................................. 1 2. Literature Review .......................................................................................................... 3 2.1 Click Fraud In Online Advertising ................................................................................... 5 2.2 Machine Learning Algorithms For Fraud Detection ....................................................... 7 3. Conceptual Model......................................................................................................... 9 4. Methodology............................................................................................................... 11 4.1 Business Understanding ............................................................................................... 11 4.1.1 Data Description ................................................................................................... 11

FAQs

What problem does the study address?

The study addresses click fraud in digital advertising, which causes financial losses and reduces advertiser trust. It focuses on detecting malicious behavior in Google Ads.

Which machine learning algorithms are used?

Five algorithms are used and compared: support vector machines, random forest, k-nearest neighbors, gradient tree boosting, and XGBoost. The work follows the CRISP-DM methodology for the project flow.

What model performs best and why?

Tree-based models perform best, particularly GTB and XGBoost. They achieve higher accuracy, recall, and AUC for identifying fraudulent clicks.

What patterns characterize fraudulent clicks in the analysis?

Non-genuine clicks are associated with primary click actions and higher click frequency per IP address and user ID. These patterns support more accurate classification and detection.

Aug 2, 2026

Upload by Miles