Ad Relevance via Keyword Clustering and Probability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current advertising systems face challenges in effectively targeting users with broad keywords that relate to multiple topics of interest, leading to a lack of relevance in ad delivery.
Innovation Solution
A system and method that determine proximity to topics by identifying match criteria associated with keywords, mapping ad keywords to clusters, computing conditional probabilities, and comparing these probabilities to assess topical proximity between search queries and advertisements, thereby enhancing ad relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If broad keywords are used in advertising, then the advertisement can reach a wider audience, but the relevance of the advertisement to specific user interests decreases
Solution Approach 1:
The patent segments broad keywords into multiple keyword clusters, each representing a distinct topic or theme. For example, a broad keyword like 'car' is divided into clusters such as 'sports cars', 'family cars', 'luxury cars', etc. This segmentation allows the system to match advertisements to specific user interests while maintaining broad coverage, as different clusters can target different user segments simultaneously.
Solution Approach 2:
The patent introduces keyword clusters as an intermediary layer between broad keywords and specific advertisements. Instead of directly matching broad keywords to ads, the system uses keyword clusters as intermediaries to bridge the gap. This intermediary structure enables more precise matching by allowing ads to be associated with specific clusters that represent particular topics, thereby improving relevance while maintaining the benefits of broad keyword coverage.
2Ease of operation
If keyword matching is simplified, then the system is easier to operate, but the precision of ad targeting decreases
Solution Approach 1:
The patent performs preliminary action by pre-computing and storing keyword clusters and their associated topics before the actual advertising matching process. The system预先 organizes keywords into clusters and calculates their topical representations in advance. This preliminary preparation simplifies the real-time operation of the system, as the matching process only needs to compare queries against pre-organized clusters rather than performing complex analysis on the fly, while still maintaining high precision through the structured cluster organization.
Data Source
AI summary
The present disclosure includes a system and method for determining proximity to topics of content items (e.g., an advertisement or “ad”). In some implementations, a method includes identifying search criteria associated with keywords of a content item. The keywords used in determining whether or not to embed the content item in Web pages including search results. Numerical scores for the search criteria is determined based, at least in part, on a topic of the content item, the numerical score indicative of proximity to the topic of the content item.