Content Targeting via User Propensity and Context Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional content delivery systems fail to effectively tailor notifications to individual user preferences, leading to reduced efficacy as they often use a one-size-fits-all approach, which may not adequately capture the varying attention-grabbing potential of different content formats for different user groups.
Innovation Solution
A system utilizing machine-learning models, including image analysis, text analysis, propensity models, and generator neural networks, to classify content objects and select subsets of users based on predicted interaction propensities, generating customized content objects that better align with user characteristics to improve engagement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the same content object is provided to all users, then the content delivery system is simple to operate, but the efficacy of the notification is reduced due to inability to capture varying attention-grabbing potential for different user groups
Solution Approach 1:
The patent segments users into different user groups based on their characteristics and preferences. The content delivery system divides the universal user base into distinct segments (e.g., users who respond to emphasized images vs. users who respond to descriptive text) and delivers tailored content objects to each segment, thereby improving notification efficacy without requiring complete system redesign
Solution Approach 2:
The patent applies local quality by customizing the content object's emphasis features according to the specific user group receiving it. Different user groups receive content with different visual or textual emphases localized to their preferences - for example, bolded text for some users, emphasized images for others - while maintaining the core notification message intact
2Productivity
If emphasized content such as bolded text or bright colors is used, then attention of users who respond to visual emphasis is improved, but users who prefer descriptive text may be less likely to respond
Solution Approach 1:
The patent makes the content object dynamic by adjusting its emphasis characteristics based on the target user group. The system dynamically selects whether to apply visual emphasis (bolded text, bright colors, emphasized images) or text-based emphasis (descriptive text) depending on the preferences of the specific user segment being targeted, allowing the same core message to adapt its presentation style
3Measurement precision
If machine-learning models are used to classify and tailor content to user groups, then the precision of user targeting is improved, but the complexity of the system increases
Solution Approach 1:
The patent implements self-service by training machine-learning models to automatically classify users into groups based on their characteristics and predict their propensity to respond to different content emphasis types. The system autonomously performs user segmentation and content customization without requiring manual intervention, thereby achieving high measurement precision while managing complexity through automation
Data Source
AI summary
Disclosed herein are techniques for machine-learning systems and methods for generating content objects using AI models. A method described herein includes predicting a propensity metric using a machine-learning propensity model describing a propensity of a user to interact with a tag. The method includes generating, using a content-tagging machine-learning model, a set of features characterizing the content object. The method includes determining, for each user in a set of users, a score that predicts a propensity of the user interacting with a particular content object. The method includes selecting a subset of users of the set of users based on the scores determined for the set of users. The method also includes facilitating output of the particular content object to each of the subset of users.