Retarget System User Score Calculation for Precision Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional retargeting systems rely on manual identification or simple rules to select users for receiving content from web services, lacking an efficient and data-driven approach to assess user value and effectiveness of content delivery.
Innovation Solution
A retarget system that calculates user scores based on their interactions with a web service, identifies high-value users, and generates data objects to deliver targeted content, using a predictive model to optimize content display and revenue generation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification or simple rules are used to select users for retargeting, then the system is easy to operate and implement, but the precision of user selection and effectiveness of content delivery is poor
Solution Approach 1:
The patent replaces manual identification and simple rule-based systems with an automated machine learning model that calculates user scores based on interaction data. This substitution enables precise user selection through algorithmic processing rather than manual or rule-based methods, directly resolving the contradiction between selection precision and system complexity.
Solution Approach 2:
The system enables self-service by automatically calculating user scores and generating retargeting lists without requiring manual intervention. The machine learning model processes interaction data autonomously to identify high-value users, eliminating the need for manual user identification while maintaining operational simplicity.
2Productivity
If a predictive model is implemented to calculate user scores and identify high-value users, then the effectiveness and relevance of content delivery is improved, but the device complexity and computational requirements increase
Solution Approach 1:
The patent implements preliminary action by pre-calculating user scores using a machine learning model trained on historical interaction data. This pre-computation enables the system to efficiently identify high-value users before content delivery, improving productivity while managing complexity through advance preparation of user segmentation.
Solution Approach 2:
The machine learning model serves as an intermediary between raw interaction data and content delivery decisions. It processes and transforms user interaction data into scored user lists that guide content delivery, mediating between complex data processing and simple delivery execution to improve effectiveness while managing system complexity.
3Measurement precision
If user interaction data is collected and processed to generate scored user lists, then the relevance of targeted content is improved, but the loss of time for data processing increases
Solution Approach 1:
The system performs preliminary processing of user interaction data to pre-calculate user scores and generate ranked user lists in advance. This pre-computation reduces the time required for real-time user selection during content delivery, resolving the contradiction between assessment accuracy and processing time by doing the work beforehand.
Data Source
AI summary
In various example embodiments, a system and method for transmitting data to select users are presented. User information corresponding to users of a web service is accessed. Scores for each of the user is calculated based on the accessed user information. Select users of the web service are identified based on the calculated scores. A data object is generated and then transmitted to the select users.