Semantic Ad Blocking System for Publisher Revenue Protection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ad blocking techniques are difficult to set up and manage, often result in under-inclusion or over-inclusion, and do not effectively work with content-targeted ad serving systems, leading to loss of advertising revenue and user dissatisfaction.
Innovation Solution
The solution involves associating concepts and terms with properties to easily block broad or narrow categories of ads, using a system that filters ads based on semantic clusters and specific terms to prevent objectionable content from being served, thereby enhancing ad blocking efficiency and relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional ad blocking techniques are used, then ads can be blocked, but the setup and management become difficult and complex
Solution Approach 1:
The patent segments ad blocking into two distinct layers: broad ad blocking based on semantic clusters and specific ad blocking based on terms. This segmentation allows each layer to handle different aspects of ad filtering independently, simplifying the overall system architecture while maintaining comprehensive blocking capability.
Solution Approach 2:
The patent introduces semantic clusters as an intermediary layer between the ad serving system and the blocking mechanism. These clusters act as mediators that group related ads together, enabling efficient broad-level blocking without requiring direct analysis of every individual ad, thus reducing system complexity.
2Ease of operation
If broad ad blocking is applied, then setup becomes easier, but under-inclusion or over-inclusion problems occur
Solution Approach 1:
The patent divides ad blocking precision into two segments: broad blocking precision handled by semantic cluster matching and specific blocking precision handled by term matching. This segmentation allows the system to achieve both ease of operation through broad blocking and high precision through specific blocking without compromise.
Solution Approach 2:
The patent applies different blocking strategies to different levels of ad classification: semantic clusters provide coarse-grained blocking for ease of operation, while individual terms provide fine-grained blocking for precision. Each level has optimized quality characteristics suitable for its purpose.
3Productivity
If content-targeted ads are served, then advertising revenue increases, but competitive ads may be displayed that harm user experience
Solution Approach 1:
The patent applies preliminary anti-action by blocking competitive ads before they are served to users. The broad and specific blocking mechanisms proactively prevent objectionable ads from entering the ad serving pipeline, eliminating the harmful effect before it can impact user experience while allowing non-competitive revenue-generating ads to proceed.
Data Source
AI summary
A content owner partner (e.g., a Website/Web page publisher) can easily block entire broad or narrow categories of ads, and can specify objectionable ad content or targeting. Concepts may be associated with a property and ads related to those concepts may be blocked for the given property. Further, terms may be associated with a property and ads including any of the terms may be blocked for the given property.