Correlated Information Recommendation via Modular Correlation Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information recommendation systems lack wide coverage and accuracy, leading to lost customers and reduced sales and browsing volume due to 'information overload' and 'information disorientation' in the vast internet space.
Innovation Solution
A method that computes a degree of correlation between visited information types and correlated information combinations using occurrence properties and user information, selecting and recommending target correlated information combinations to improve recommendation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing recommendation systems use simple product or information matching, then the system complexity is low, but the recommendation accuracy and coverage are insufficient
Solution Approach 1:
The patent segments the recommendation process into multiple independent modules: user behavior data collection, correlation rule generation, correlation degree computation, and recommendation output. Each module handles a specific aspect of the recommendation task, allowing the system to achieve high accuracy through modular processing without becoming unmanageably complex.
Solution Approach 2:
The patent transitions from traditional single-item recommendation to multi-dimensional correlation analysis by introducing correlation rules that consider multiple information types (product, category, attribute, user behavior) simultaneously. This dimensional expansion enables more accurate recommendations while maintaining systematic organization through structured correlation rules.
2Adaptability or versatility
If the recommendation system analyzes more information types and user behaviors, then the recommendation coverage improves, but the data processing complexity increases
Solution Approach 1:
The patent creates a universal correlation rule structure that can handle multiple information types (products, categories, attributes, user behaviors) through a single standardized framework. The correlation rule template is designed to be multi-functional, accommodating different data types without requiring separate processing mechanisms for each, thus expanding coverage while controlling complexity.
Solution Approach 2:
The system manages complexity by dynamically adjusting correlation degree parameters based on different information types and user contexts. Rather than processing all possible correlations equally, the system modifies correlation parameters to focus computational resources on the most relevant relationships, expanding effective coverage while limiting unnecessary processing complexity.
3Measurement precision
If the system computes correlation degrees based on multiple occurrence properties, then the recommendation precision improves, but the computational time increases
Solution Approach 1:
The patent pre-computes and stores correlation rules and their associated parameters during off-peak periods or during data collection phases. By performing preliminary analysis and storing intermediate results, the system avoids re-computing all correlations in real-time, thus maintaining high precision while reducing actual recommendation delivery time.
Solution Approach 2:
The system computes correlation degrees for only the most relevant information combinations based on user context and behavior patterns, rather than calculating all possible correlations. This partial computation approach focuses computational effort on the subset of correlations most likely to influence the recommendation, achieving sufficient precision with reduced computational time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for information recommendation uses correlated information combinations to improve recommendation accuracy. Upon receiving data from a client indicating a visited information type, the method obtains correlated information types related to the visited information type from stored records. The correlated information types providing one or more correlated information combinations each including at least two correlated information types. For each of the one or more correlated information combinations, the method computes a degree of correlation between the correlated information combination and the visited information type. The method selects a target correlated information combination with a satisfying degree of correlation, and recommends the target correlated information combination to the client.