Click Fraud Detection in Online Advertising: A Comparative Study of Machine Learning Models
Bibliographic record
Abstract
Advancements in networking and communication technologies have significantly boosted digital advertising, with global spending expected to reach $646 billion by 2024, including $495 billion from mobile internet. However, this growth is hindered by the persistent issue of click fraud, which leads to substantial financial losses and distorts advertising metrics. This study presents a comprehensive comparative analysis of multiple machine learning (ML) models including Random Forest, LightGBM, XGBoost, AdaBoost, Decision Tree, Gradient Boosting, and Multi-Layer Perceptron (MLP), for detecting click fraud in online advertising. A key novelty of this work lies in the integration of the LIME (Local Interpretable Model-agnostic Explanations) framework, which enhances transparency by interpreting the decision-making process of complex models. Through extensive data preprocessing and model evaluation using metrics such as accuracy, precision, recall, and F1-score, the Random Forest model achieved the highest accuracy of 95%, demonstrating robustness and generalization across different scenarios. Unlike prior studies, this research emphasizes model interpretability and trustworthiness, providing actionable insights for advertisers and platform designers. Comparative analysis with existing literature further highlights the methodological effectiveness and practical relevance of the proposed approach.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.028 | 0.048 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.002 |
| Bibliometrics | 0.007 | 0.004 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.003 | 0.005 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".