Ad Serving System Using Content Analysis for Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic ad serving systems, such as Google Adwords, often present ads that are not relevant to users' search queries due to keyword bidding, leading to frustrating experiences and inefficient ad delivery, as they do not consider the content of retrieved search results when selecting ads.
Innovation Solution
A system that identifies topics from search results and uses a combination of keyword targeting and content analysis to select and present ads within the search interface, allowing for more relevant ad placement by considering the content of retrieved web pages and user interactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword bidding is used to select ads, then advertisers can compete for user attention, but ads become irrelevant to users' search queries
Solution Approach 1:
The patent introduces an intermediary system that analyzes the content of search result webpages and uses this analysis to select ads. The system acts as a mediator between keyword bidding and ad selection, using webpage content analysis as an intermediate step to determine ad relevance, thereby resolving the contradiction between competitive bidding and ad relevance.
Solution Approach 2:
The system implements feedback by analyzing user interactions with search results and using this information to improve ad selection. The patent describes using user behavior data and search result content to continuously refine ad relevance, creating a feedback loop that addresses the irrelevance problem while maintaining bidding efficiency.
2Loss of energy
If ads are selected based on keyword bidding, then advertising revenue can be maximized, but user experience deteriorates due to irrelevant ads
Solution Approach 1:
The patent changes the parameters used for ad selection from purely keyword-based to a hybrid model that incorporates webpage content analysis. By changing the selection parameters to include content relevance scores alongside bidding information, the system maintains revenue maximization while improving user experience through more relevant ads.
Solution Approach 2:
The ad selection system uses a composite approach, combining multiple factors including keyword bids, webpage content analysis, and user interaction data. This composite methodology creates a more robust ad selection mechanism that balances revenue goals with user experience, similar to how composite materials combine different properties to achieve superior performance.
3Reliability
If content analysis of search results is performed, then ad relevance improves, but system complexity increases
Solution Approach 1:
The patent applies segmentation by dividing the ad selection process into distinct modules: keyword bidding analysis, webpage content analysis, user interaction tracking, and ad ranking. This segmentation allows the system to handle complexity through modular design, where each component can be independently optimized and maintained, thereby improving ad relevance without overwhelming system complexity.
4Reliability
If multiple factors are considered for ad selection, then ad performance improves, but processing time increases
Solution Approach 1:
The system performs preliminary action by pre-analyzing webpage content and caching relevance information before actual ad selection occurs. By preparing content analyses in advance and storing them for quick retrieval, the system can consider multiple factors during ad selection without significantly increasing processing time, thus improving ad performance while minimizing time loss.
Data Source
AI summary
A system and method are disclosed for presenting relevant advertising to user search queries. The ads can be selected by reference to an underlying content presented in organic search results. The ads can be combined and supplemented with conventional keyword/pay for placement type services to improve ad performance. The system can also be used for testing content related ads in advance of their presentation to potential third party partners in an advertising campaign.