Deep Learning-based Model to Fight Against Ad Click Fraud

Abstract

Click fraud is a fast-growing cyber-criminal activity with the aim of deceptively clicking on the advertisements to make the proit to the publisher or cause loss to the advertiser. Due to the popularity of smartphones since the last decade, most of the modern-day advertisement businesses have been shifting their focus toward mobile platforms. Nowadays, in-app advertisement on mobile platforms is the most targeted victim of click fraud. Malicious entities launch attacks by clicking ads to artiicially increase the click rates of speciic ads without the intention of using them for legitimate purposes. The fraud clicks are supposed to be caught by the ad providers as part of their service to the advertisers; however, there is a lack of research in the current literature for addressing and evaluating different techniques of click fraud detection and prevention. Another challenge toward click fraud detection is that the attack model can itself be an active learning system (smart attacker) with the aim of actively misleading the training process of fraud detection model via polluting the training data. In this paper, we propose a deep-learning based model to address the challenges as mentioned above. The model is a hybrid of artiicial neural network (ANN), auto-encoder and semi-supervised generative adversarial network (GAN). Our proposed approach triumphs excellent accuracy than other models. CCS CONCEPTS· Information systems → Online advertising; · Security and privacy → Intrusion/anomaly detection and malware mitigation; · Computing methodologies → Machine learning.

Mini-Batch Normalized Mutual Information: A Hybrid Feature Selection Method

“…In the first version, we considered one million rows of data in which the ratio of classes match the ratio at 200 million rows (Talkingdata Version 1) is taken. 913692 data samples were used for the second variant, where the rows were equally categorized into two classes (Talkingdata Version 2) [29].…”

GFD: A Weighted Heterogeneous Graph Embedding Based Approach for Fraud Detection in Mobile Advertising

“…To automatically detect mobile advertising fraud behaviors, machine learning methods have been successfully applied to find fraud patterns in data, distinguishing suspicious advertising fraud operation from normal one [10][11][12][13][14]. As for learning model with attribute features, researchers usually use several attributes from each sample to train a learning model to identify the fraud behaviors.…”

CKMI: Comprehensive Key Management Infrastructure Design for Industrial Automation and Control Systems

“…The communication messages contain sensitive data; if an attacker captures or modifies the message contents, it results in severe consequences [4,7]. The following are some of the malicious actions that the attacker may perform:…”

“…Generate 128-bit field level group key (FGK) 7: else if Key generation is for control level devices then 8: Choose matrix method for key establishment 9: Generate secret symmetric matrix S and public matrix P 10:…”

“…5: else if Pair-wise key is lost between plant devices then 6: Devices regenerate it by exchanging public keys and by performing scalar multiplication operation. 7: else if All the security parameters of the device are lost then 8: Key recovery is done by CKMI server using the commissioning device 9: end if 4.10. Key de-Registration and destruction CKMI server removes all the traces of keying materials from local storage, backup and archive when the usage of keying materials is no longer required.…”