N-Tree Model for Friend Recommendation Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing social network service (SNS) technologies face challenges in accurately expanding relationship chains and recommending friends due to empirically configured and fixed weight values for common relationship factors, which limits the accuracy of friend recommendation.
Innovation Solution
A method and device utilizing a pre-configured N-Tree prediction model, based on gradient boosting decision trees, to process feature data from user associations, automatically determining association-predicting values for friend recommendations by learning to generate weight values for relationship factors, thereby facilitating intelligent and automated friend recommendation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If empirically identified and artificially configured weight values are used for common relationship factors, then the configuration process is simple, but the friend recommendation accuracy deteriorates
Solution Approach 1:
The system enables self-service by allowing the model to automatically learn and determine optimal weight values for relationship factors through training on user interaction data, eliminating the need for manual empirical configuration while achieving high recommendation accuracy
Solution Approach 2:
The invention transforms fixed empirically configured parameters into dynamic parameters that are automatically adjusted through machine learning. The weight values for relationship factors are no longer static but are learned and optimized based on actual user behavior patterns, enabling the system to adapt to changing user preferences
2Ease of operation
If fixed weight values are used for relationship factors, then the system is easy to implement, but the adaptability to different user scenarios deteriorates
Solution Approach 1:
The system transitions from static fixed weight values to dynamic adaptive weights that automatically adjust based on user interactions and contextual factors. The model continuously learns from new data, enabling it to adapt to different user scenarios and evolving relationship patterns without requiring system reconfiguration
Solution Approach 2:
The invention incorporates feedback mechanisms where user interactions with recommended friends are used to continuously refine and update the weight values for relationship factors. This closed-loop feedback system enables the model to learn from actual outcomes and improve its adaptability to diverse user scenarios over time
3Device complexity
If manual configuration of weight values is performed, then the system complexity is low, but the automation level deteriorates
Solution Approach 1:
The invention replaces the manual mechanical configuration process with an automated machine learning system. Instead of manually setting weight values through empirical judgment, the system uses algorithms to automatically learn optimal weights from data, substituting human cognitive processes with computational processes that achieve higher automation and accuracy
Data Source
AI summary
Methods, devices, and servers for friend recommendation are provided. A user association set of a target user is obtained. Original data of each associated user in the user association set is obtained. The original data include location relationship data, associated friend data, time relationship data, or combinations thereof, between each associated user and the target user. The original data of each associated user is screened to obtain feature data to form a feature collection for each associated user. A pre-configured N-Tree prediction model is used to process the feature collection for a prediction calculation to obtain an association-predicting value for each associated user. According to the association-predicting value of each associated user, a friend user for the target user from the user association set is determined and recommended to the target user.