Ad Fraud Detection via Distributed Data Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the digital advertisement market, it is challenging to verify malicious acts between advertisers, advertisement platforms, and advertisement media due to the lack of trust and the numerous advertisement media, making it difficult to detect fraudulent activities such as click spamming, click injection, and fake installs.
Innovation Solution
An electronic device and method for sharing data between devices to verify advertisement-related information, using a program that includes encryption and an AI model trained on fraud data to verify the reliability of shared data, ensuring trust between advertisers, platforms, and media by generating and comparing event data and identifying fraudulent activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If service is provided based on trust between advertiser, advertisement platform, and advertisement media, then the system operation is simple, but malicious acts cannot be verified and fraud detection is difficult
Solution Approach 1:
The patent applies preliminary action by having electronic devices share verification programs and event data before fraudulent activities can occur. The system proactively establishes trust relationships and verification mechanisms in advance, allowing devices to verify each other's authenticity and detect potential fraud before it impacts the advertising ecosystem.
Solution Approach 2:
The patent introduces an intermediary verification mechanism where event data serves as a mediator between advertiser, advertisement platform, and advertisement media. This intermediary data structure allows indirect verification of fraudulent activities by comparing event information across multiple parties without requiring direct trust relationships between them.
2Reliability
If the number of advertisement media is increased, then the market coverage is expanded, but detecting malicious acts by multiple advertisement media becomes difficult
Solution Approach 1:
The patent applies segmentation by dividing the verification process into individual event data units that can be independently verified. Each advertisement media generates separate event data records for specific advertising events, allowing the system to detect fraudulent patterns in individual media without being overwhelmed by the complexity of verifying all media simultaneously.
Solution Approach 2:
The patent implements feedback mechanisms where event data is continuously shared and verified across the network. When fraudulent activity is detected in one advertisement media, the verification program uses this feedback information to identify similar patterns in other media, enabling scalable fraud detection that improves accuracy as more data becomes available.
3Reliability
If data sharing between electronic devices is implemented, then trust verification is enabled, but data security and privacy protection become challenging
Solution Approach 1:
The patent extracts only the necessary verification information from event data and shares it across devices through the verification program. Instead of sharing complete raw data sets that could compromise security, the system extracts specific event metrics and verification tokens that are sufficient for trust verification while maintaining data privacy and security.
Data Source
AI summary
Disclosed are an electronic device and a method for controlling same. A method for controlling an electronic device according to the present disclosure comprises: a step of obtaining a program which shares data about an advertisement with another electronic device so as to verify the shared data; a step of, when an event for the advertisement occurs, generating first data including information about the event for the advertisement; a step of transmitting the generated first data to the other electronic device; a step of receiving second data including information about an event from the advertisement generated from the other electronic device; and a step of verifying the second data using the program.