Recommendation Ranking System with Distrust Integration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing recommendation ranking systems fail to fully utilize negative trust information, leading to inaccurate recommendations, as they either ignore distrust or do not effectively offset positive and negative expressions of trust.
Innovation Solution
A method to incorporate negative trust information into the ranking system by computing consistent trust values for members of a social network, where opinions from trusted members are weighted more heavily and those with low trust are discounted, ensuring that no individual has a disproportionate impact on the overall recommendation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If negative trust information is completely ignored in the ranking system, then the system maintains simplicity and avoids over-complication, but the accuracy of recommendations deteriorates because distrust information cannot offset positive trust information
Solution Approach 1:
The patent segments trust information into positive trust and negative trust components, allowing them to be processed separately and then combined. This segmentation enables the system to handle distrust information without completely restructuring the ranking system, thus improving accuracy while controlling complexity.
Solution Approach 2:
The patent introduces a trust parameter that can take positive or negative values, changing the traditional single-dimensional trust metric. This parameter change allows the system to represent both trust and distrust quantitatively, enabling more accurate recommendations through mathematical combination of opposing trust signals.
2Reliability
If all member opinions are weighted equally in the collaborative ranking system, then the system maintains simplicity and equal participation, but the quality of recommendations deteriorates because opinions from untrusted members have the same impact as trusted members
Solution Approach 1:
The patent applies local quality by assigning different weights to different members based on their individual trust levels. Instead of uniform weighting, each member's opinion is weighted according to their specific trust score, allowing the system to emphasize opinions from trusted members while diminishing those from untrusted members, thus improving recommendation quality.
3Measurement precision
If the system incorporates comprehensive trust propagation through multiple network levels, then the accuracy of trust assessment improves, but the computational complexity and time required for ranking increases significantly
Solution Approach 1:
The patent implements partial action by limiting trust propagation to a specified number of network levels rather than computing trust across the entire network indefinitely. This partial computation approach provides sufficiently accurate trust assessments for practical purposes while significantly reducing the computational time and resources required compared to exhaustive multi-level propagation.
Data Source
AI summary
A recommendation ranking system that computes trust for entities based on negative expressions of trust. Negative expressions of trust are used to reduce the trust of entities. However, the system may discount entities that in the aggregate are distrusted. The system may be used with a social network to provide accurate, personalized recommendations for members of the social network. The network may be modeled as a voting network, with each member of the social network represented as a node and expressions of trust between members represented as weights on edges between nodes. Values of trust may be computed for nodes in the network and used to generate a recommendation. Opinions expressed on a topic may be weighted by trust in the node expressing the opinion. The system may be applied in other settings that can be modeled as a voting network, including ranking of Internet search results.