Dynamic Ad Selection Threshold Based on Score Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing online advertisement selection methods often display advertisements with quality scores just above a threshold, resulting in a collection that may appear marginally relevant to users, as they do not consider the relationships among multiple advertisement candidates, leading to a lack of overall advertisement quality.
Innovation Solution
A method that calculates a score threshold based on the relationships among multiple quality scores of advertisement candidates, identifying and removing candidates with low scores to ensure a subset of advertisements that collectively meet a predetermined measure of quality, and replacing them with higher-scoring candidates, thereby enhancing the overall quality of displayed advertisements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If advertisement candidates with quality scores just above a threshold are displayed, then the number of displayed advertisements increases, but the overall quality and relevance of the advertisement collection deteriorates
Solution Approach 1:
The patent changes the threshold parameter from a fixed value to a dynamic value that is calculated based on the relationship among multiple quality scores. The threshold is adjusted according to the distribution of scores in the candidate pool, ensuring that the displayed advertisements maintain high overall quality while still achieving sufficient quantity.
Solution Approach 2:
The patent performs preliminary evaluation and ranking of all advertisement candidates before selection. By pre-calculating quality scores and establishing the threshold based on the score distribution, the system ensures that only advertisements meeting the quality criteria are considered for display, maintaining collection quality while maximizing quantity.
2Productivity
If a fixed quality score threshold is used for selection, then the selection process is simple and fast, but the advertisement collection lacks cohesion and overall quality
Solution Approach 1:
The patent transforms the static threshold parameter into a dynamic one that adapts to the quality score distribution of candidate advertisements. This allows the system to maintain selection efficiency while ensuring that the threshold reflects the actual quality level of available candidates, thereby improving collection cohesion.
Solution Approach 2:
The threshold calculation incorporates feedback from the quality scores of multiple candidates. By analyzing the distribution and relationships among scores, the system adjusts the threshold to ensure that the selected advertisements form a cohesive, high-quality collection rather than merely meeting an arbitrary fixed standard.
Data Source
AI summary
This specification describes technologies relating to displaying online content. In general, one aspect of the subject matter described in this specification can be embodied in methods that include receiving a collection of advertisement candidates for display in an online medium, the advertisement candidates each assigned a quality score calculated based at least in part on a measure indicative of relevance of the respective advertisement candidate to online content for concurrent display in the online medium, determining a score threshold based at least in part on relationships among multiple quality scores of the quality scores associated with the advertisement candidates in the collection of advertisement candidates, and based on the determined score threshold, identifying a subset of advertisement candidates of the collection for display. Other embodiments of this aspect include corresponding systems, apparatus, and computer program products.