Ad Delivery System Integrating Keyword and Content Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing electronic ad serving systems, such as Google Adwords, often present irrelevant ads to users due to keyword bidding that is not related to the user's query or the content of the webpage, leading to frustrating experiences and inefficient ad delivery.
Innovation Solution
A system that integrates keyword targeting with content analysis to select relevant ads within the search page, using a blend of keyword and content-based methods to identify and present ads that are more closely aligned with user interests, by analyzing the topics of retrieved search results and presenting content-related ads alongside keyword ads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If keyword bidding is used to select ads, then advertisers can compete for user attention, but irrelevant ads are presented to users leading to poor user experience
Solution Approach 1:
The patent combines keyword-based advertising with content-based advertising by analyzing the content of search result pages and presenting ads that are relevant to both the search query keywords and the actual content displayed. This merging of approaches ensures that ads are selected based on multiple factors including keyword relevance and content relevance, thereby improving both ad delivery efficiency and ad relevance simultaneously
Solution Approach 2:
The patent introduces content analysis as an intermediary mechanism between keyword bidding and ad selection. By analyzing the content of search result pages and identifying relevant topics, the system acts as a mediator that filters and refines the ad selection process, ensuring that only ads relevant to both keywords and content are presented to users
2Productivity
If ads are presented based solely on keyword bidding, then advertisers can maximize visibility, but ad relevance to user interests deteriorates
Solution Approach 1:
The patent segments the ad selection process into multiple independent components: keyword-based selection, content-based selection, and hybrid selection. By analyzing search result page content separately from keyword bids, the system can independently optimize for visibility and relevance, then combine the results to present ads that satisfy both requirements
Solution Approach 2:
The patent applies different selection criteria to different aspects of ad presentation. Keyword bidding determines overall ad visibility and competition, while content analysis determines local relevance and specificity. This local quality approach ensures that ads are optimized for both visibility and relevance to user interests based on the specific content context
3Reliability
If content analysis is added to ad selection, then ad relevance improves, but system complexity increases
Solution Approach 1:
The patent performs content analysis of search result pages as a preliminary action before ad selection. By pre-analyzing the content and identifying relevant topics in advance, the system prepares relevant ad candidates that can be quickly matched with user interests and keywords, reducing the overall complexity of real-time ad selection while maintaining high relevance
Data Source
AI summary
A system and method is disclosed for presenting relevant advertising to user search queries. The ads are based on content which is derived from a set of documents/pages from websites forming a collective.