Social Annotation Construction for Ad Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Internet-based advertising, particularly banner ads, often fail to resonate with users as they are not tailored to individual interests, leading to low click-through rates and missed opportunities for advertisers.
Innovation Solution
A method that involves receiving data on user conversions associated with social contacts, filtering relevant records, and constructing detailed annotations to enhance ad relevance by incorporating user IDs and conversion actions, which are then sent to clients, boosting ad scores based on affinity scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If generic banner ads are shown to large audiences, then advertising reach is improved, but ad relevance deteriorates
Solution Approach 1:
The patent segments the generic advertising audience into specific user groups based on their social connections. By dividing the broad audience into smaller segments with shared characteristics (friends who performed similar actions), the system maintains large-scale reach while improving relevance for each segment through personalized annotations.
Solution Approach 2:
The patent applies local quality by adding personalized annotations to ads based on the specific characteristics of each user's social network. Instead of uniform generic ads, each user receives ads with locally tailored information about their friends' actions, making the ad content relevant to their specific social context while maintaining broad distribution.
2Loss of information
If ads are tailored to individual user interests, then ad relevance is improved, but system complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing annotations that describe friends' actions and interests before ad delivery. These annotations are prepared in advance and cached, so when an ad is delivered, the system can quickly attach relevant personalized information without performing complex real-time analysis, thus reducing operational complexity while maintaining high relevance.
Solution Approach 2:
The patent uses copying by creating standardized annotation templates that can be reused across multiple users. Instead of generating unique complex analyses for each user, the system copies and adapts pre-defined annotation patterns based on user-specific data, significantly reducing the computational complexity of personalization while maintaining individual relevance.
3Adaptability or versatility
If detailed conversion records are retrieved and filtered for each user, then ad personalization is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-filtering and organizing conversion records into user-specific annotation sets before ad delivery. The system prepares personalized annotation data in advance based on social connection data, so when an ad request comes in, the filtering and matching process is significantly accelerated because the relevant data is already organized and ready for quick retrieval.
4Adaptability or versatility
If social connection data is integrated into ad delivery, then user appeal is improved, but data processing requirements increase
Solution Approach 1:
The patent applies the extraction principle by separating social connection data processing from the main ad delivery workflow. The system extracts and processes social network information independently to create annotation profiles, then integrates these pre-processed annotations into ad delivery. This separation reduces the real-time data processing burden during ad serving while still incorporating valuable social connection insights for improved user appeal.
Data Source
AI summary
Methods, apparatuses, and computer-readable media for serving detailed annotations is disclosed. When a user converts in response to an advertisement, data about the conversion is stored in one or more logs. A batch handler uses the one or more logs to create and enter a record regarding the conversion into a conversion table. Upon receiving an ad request from a client, ads scorer may construct a detailed annotation using the conversion table and a social graph table, and ads server may return an ad and the detailed annotation. Ads server may also return an ad plus an iframe. The iframe requests a detailed annotation from a conversion management server. The conversion management server constructs and returns the detailed annotation.