Sequence-Based Content Pushing for Real-Time Recommendation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing content pushing technologies based on artificial intelligence fail to align with real-time user needs, resulting in low accuracy and resource wastage.
Innovation Solution
A content pushing method that utilizes sequence-based feature encoding and extraction from user viewing history to select and push personalized content, incorporating a neural network model for encoding and selecting candidate content based on encoded features and related features of current and historical viewed content.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If historical viewing features are used for content pushing, then personalization is achieved, but real-time accuracy deteriorates
Solution Approach 1:
The patent transforms the static historical feature analysis into a dynamic sequence processing system. By treating user viewing history as a time-ordered sequence and applying sequential encoding, the system adapts to changing user preferences in real-time while maintaining personalization capabilities.
Solution Approach 2:
The patent introduces a temporal dimension to the feature analysis by processing viewing history as a sequence rather than aggregated statistics. This sequential dimension allows the system to capture evolution of user preferences over time, improving real-time accuracy while preserving personalization.
2Productivity
If traditional AI analysis is used, then computing resources are consumed, but pushing accuracy remains low
Solution Approach 1:
The patent extracts only the essential sequential patterns from user viewing history that are most relevant to real-time recommendations. By focusing on key temporal features rather than processing all historical data equally, the system reduces computational overhead while improving recommendation accuracy.
Solution Approach 2:
The patent changes the parameter representation from aggregated historical statistics to sequential feature vectors that capture temporal dynamics. This parameter transformation enables more efficient processing with reduced resource consumption while achieving higher pushing accuracy through better representation of user intent.
Data Source
AI summary
A content pushing method, apparatus, and computer-readable storage medium for personalized content recommendation based on viewing sequence analysis. The method obtains a user's current information sequence containing information items arranged in viewing order, including current and historical viewed content. The sequence is encoded into feature sequences with corresponding feature items. Description information of current viewed content is encoded to obtain its encoded feature. Related features are extracted from the feature sequence based on the current viewed content's encoded feature. Description information of candidate content is encoded to obtain encoded features for each candidate. A candidate content is selected based on comparison between candidate encoded features and the related feature of current viewed content, then pushed to the user for personalized recommendations.