Advertising Activity Detection via Browser and User Data Discrepancy Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
On-line advertising systems are vulnerable to malicious activities such as click-bots and impression-bots that inflate advertising link click and impression counts, leading to inaccurate fee calculations for companies and increased revenue for web publishers.
Innovation Solution
A system that collects and compares information during advertising link requests and clicks to detect illegitimate activity by matching browser, client, and user information, determining illegitimate activity when discrepancies are found.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If click-bots and impression-bots are used to inflate advertising link click and impression counts, then the advertising system revenue increases, but the accuracy of fee calculations deteriorates
Solution Approach 1:
The system continuously monitors advertising link interactions and compares expected behavior patterns against actual behavior. When bots attempt to inflate metrics, their abnormal patterns are detected through feedback loops that compare click sequences, timing patterns, and user behavior consistency, allowing the system to identify and exclude fraudulent activity from fee calculations.
Solution Approach 2:
The patent replaces simple mechanical counting of clicks and impressions with a sophisticated software-based detection system that analyzes behavioral patterns, user profiles, and interaction sequences. This substitution of mechanical measurement with intelligent analysis enables the system to distinguish between legitimate users and bots, maintaining fee calculation accuracy while preserving revenue from genuine advertising engagement.
2Device complexity
If simple click tracking is used, then the system complexity is low, but the ability to detect illegitimate activity deteriorates
Solution Approach 1:
The detection system segments the advertising interaction process into multiple observable components: user profile data, click sequence patterns, timing intervals, and navigation behavior. By dividing the complex detection task into these discrete segments, the system can analyze each component independently using appropriate methods, making the overall detection capability more manageable and scalable while improving bot identification accuracy.
Solution Approach 2:
The system transitions from single-dimensional click counting to multi-dimensional analysis by incorporating temporal patterns, spatial relationships between ad elements, user demographic dimensions, and sequential behavior patterns. This dimensional expansion creates a richer detection space where bot behavior stands out more clearly against legitimate user patterns, enhancing detection capability without proportionally increasing system complexity.
3Measurement precision
If comprehensive information collection is implemented, then the detection accuracy of illegitimate activity improves, but the information processing requirements increase
Solution Approach 1:
The system performs preliminary actions by pre-establishing user profiles, expected behavior patterns, and baseline statistical distributions during normal operation. This advance preparation allows the system to detect illegitimate activity through simple comparison operations rather than complex real-time analysis, reducing processing resource requirements while maintaining high detection accuracy. The heavy lifting is done beforehand when data is abundant and less critical.
Solution Approach 2:
The system dynamically adjusts the scope and depth of information collection based on detected anomalies and operational context. Rather than consistently gathering all possible data, the system modifies its parameter set - collecting comprehensive information only when needed for detection, and using simplified tracking for normal operations. This adaptive parameter adjustment balances detection accuracy with processing resource consumption, allocating computational effort proportionally to risk and complexity.
Data Source
AI summary
A system may obtain a first group of information when an advertising link is requested and a second group of information when the advertising link is selected. The system may further detect illegitimate advertising-related activity based on the first group of information and the second group of information.