Ad Relevance via Search Result Content Analysis
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 and method that integrates keyword targeting with content analysis to select and present ads within a search interface, using techniques like term vector databases and synthetic pages to identify relevant ads based on the content of search results, allowing for more relevant ad placement and improved user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword-based ad bidding is used, then advertisers can compete for user attention, but ads may not be relevant to users' search queries
Solution Approach 1:
The patent combines keyword-based ad bidding with content-based ad selection by analyzing the content of retrieved search results. The system merges the competitive bidding mechanism with content analysis to select ads that are both competitively bid-for and relevant to the actual search query and result content, thereby resolving the contradiction between ad delivery efficiency and ad relevance.
Solution Approach 2:
The patent introduces content analysis as an intermediary factor between keyword bidding and ad selection. Instead of directly selecting ads based solely on keyword bids, the system uses content analysis of search results as a mediator to filter and refine ad selections, ensuring that ads are relevant to users' actual information needs while still allowing keyword-based competition.
2Ease of operation
If ads are presented based on keyword bidding, then advertisers can secure prominent positions, but users experience frustration from irrelevant ads
Solution Approach 1:
The patent implements feedback by analyzing the content of retrieved search results and using this information to adjust ad selections. The system continuously refines ad choices based on the actual content returned, ensuring that ads remain relevant to users' search intentions and reducing frustration from irrelevant advertising.
Solution Approach 2:
The patent changes the parameters for ad selection by incorporating content-based factors alongside keyword bidding parameters. The system adjusts ad selection criteria to include content relevance metrics, thereby reducing the harmful effect of irrelevant ads while maintaining the effectiveness of keyword-based advertising.
3Reliability
If content analysis is added to ad selection, then ad relevance improves, but system complexity increases
Solution Approach 1:
The patent segments the ad selection process into distinct components: keyword bidding evaluation, content analysis of search results, and ad matching based on both factors. This segmentation allows the system to handle complexity systematically by processing different aspects of ad selection separately and then integrating them, thereby improving ad relevance while managing system complexity through modular architecture.
Data Source
AI summary
A system and method is disclosed for presenting relevant advertising to web pages, including web logs. The ads can be selected by reference to an underlying content presented in web logs, based on extracting topics from the pages, and using these topics as the basis for a search query. Ads can then be matched against another set of topics derived from web pages forming the search results to the query.