Resource Pushing via Channel and Content Preference Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing resource pushing methods in AI technologies face challenges in improving click-through rates (CTRs) due to limited consideration of user preferences, resulting in poor resource pushing effects and low CTRs, as they primarily sort resources based on predicted CTRs without integrating channel and content preferences effectively.
Innovation Solution
A resource pushing method that integrates channel and content preferences by using a target recommendation model to obtain target resources, which are then pushed to users, enhancing the personalization of resource recommendations and improving CTRs by considering both channel and content preferences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If candidate resources are sorted directly according to predicted CTRs, then the sorting process is simple and fast, but the resource pushing effect is poor and CTRs are low
Solution Approach 1:
The patent segments the resource sorting process into multiple stages: initial CTR prediction, preference feature extraction (channel and content preferences), and multi-dimensional scoring. This segmentation allows each stage to focus on specific aspects, improving overall recommendation quality while maintaining efficiency.
Solution Approach 2:
The patent introduces additional dimensions beyond CTR prediction by incorporating channel preference features and content preference features. This multi-dimensional approach transforms the single-dimensional CTR sorting into a comprehensive evaluation system that considers user preferences across different channels and content types.
2Device complexity
If only CTR prediction is used for resource sorting, then the model complexity is low, but user preferences in different dimensions are not effectively integrated
Solution Approach 1:
The preference modeling is segmented into distinct components: channel preference features and content preference features. Each component is processed separately through dedicated neural network layers, allowing the system to capture different aspects of user preferences independently before combining them for final resource selection.
Solution Approach 2:
The patent creates a composite preference representation by combining channel preference features and content preference features into a unified preference vector. This composite structure integrates multiple types of user preferences, enabling more versatile and personalized resource recommendations.
3Productivity
If limited information is taken into consideration in resource sorting, then the processing is efficient, but the personalization of recommendations is insufficient
Solution Approach 1:
The patent performs preliminary extraction of preference features from user historical data before the main resource sorting process. By pre-computing channel preferences and content preferences from user behavior patterns, the system prepares personalized preference profiles in advance, enabling efficient and accurate recommendation matching during resource pushing.
Data Source
AI summary
This application discloses a resource pushing method performed by a computer device. The method includes: obtaining a target recommendation model and a preference feature and a candidate resource set corresponding to a target object, the preference feature including at least a channel preference feature and a content preference feature; obtaining at least one target resource from the candidate resource set based on the target recommendation model and the preference feature; and pushing the at least one target resource to the target object. Such a resource pushing process integrates preferences of the target object in different dimensions, so that the target resource pushed to the target object not only conforms to channel preferences of the target object, but also conforms to content references of the target object, which is beneficial to improving the resource pushing effect, and further increasing the click-through rates (CTRs) of the pushed resources.