Web Page Risk Rating for Brand Safety in Online Advertising
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current online advertising systems lack efficient methods to identify and control the placement of advertisements on web pages containing objectionable content, leading to misalignment of brand images and incentives in the ad-serving ecosystem.
Innovation Solution
Developing mechanisms to score and rate web pages and content using ordinal models that generate risk ratings based on the probability and severity of objectionable content, allowing advertisers to make informed decisions about ad placement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current ad-placement processes are used, then advertisements can be served on web pages, but advertisements may appear alongside objectionable content due to opacity and misalignment of incentives
Solution Approach 1:
The patent introduces an intermediary rating system that acts as a mediator between advertisers and web pages. Independent raters assess web pages and assign ratings that are visible to advertisers, thereby mediating the ad-placement process and enabling informed decisions without requiring direct inspection of all web page content
Solution Approach 2:
The patent implements feedback mechanisms where rating information about web pages is provided back to advertisers and ad networks. This feedback loop enables continuous improvement of ad-placement decisions by incorporating quality assessments into the bidding and selection processes
2Reliability
If advertisers want to control ad placement on safe web pages, then brand safety improves, but the complexity of the ad-placement process increases
Solution Approach 1:
The patent applies preliminary action by having web pages rated in advance before advertisements are placed on them. Rating information is pre-computed and made available to advertisers, eliminating the need for complex real-time content analysis during the ad-placement process
Solution Approach 2:
The patent transforms the complex quality assessment problem into a simplified parameter-based system. Instead of evaluating multiple content quality dimensions in real-time, the system uses pre-computed rating parameters that advertisers can directly apply as filtering or weighting criteria
3Measurement precision
If comprehensive content analysis is performed to identify objectionable content, then measurement precision improves, but the time and computational resources required increase
Solution Approach 1:
The patent performs content analysis and rating computation in advance, before the ad-serving process. Web pages are assessed by independent raters and assigned ratings that can be cached and reused for multiple ad-auction decisions, avoiding repeated analysis of the same content
Solution Approach 2:
The patent segments the content assessment task into independent rating components that can be performed by specialized raters. Different aspects of content quality are evaluated separately and aggregated into composite ratings, enabling parallel processing and reducing overall computation time
Data Source
AI summary
Systems, methods, and media for rating websites for safe advertising are provided. In accordance with some embodiments of the disclosed subject matter, the method comprises: extracting one or more features from a piece of web content; applying a plurality of statistical models to the extracted features to generate a plurality of ordinomial estimates, wherein each ordinomial estimate represents a probability that the web content is a member of one of a plurality of severity groups; determining a posterior ordinomial estimate for the web content by combining the plurality of ordinomial estimates; generating a risk rating that encodes severity and confidence based on the determined posterior ordinomial estimate, wherein the risk rating identifies whether the web content is likely to contain objectionable content of a given category; and providing the risk rating for determining whether an advertisement should be associated with the web content.