User Portrait Matrix Generation via Automated Tag Weight Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional user portrait obtaining methods require significant time and labor to manually set tag weights, leading to high costs and low accuracy due to potential errors in manually setting tag weights.
Innovation Solution
A method that uses training samples from user behavior logs to modify initialized user and tag parameter matrices through a data fitting model, generating a user portrait matrix that represents preference degrees without the need for manual tag weight setting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If manual tag weight setting is used, then flexibility in adjusting user portrait parameters is improved, but time and labor costs increase significantly
Solution Approach 1:
The system automatically determines tag weights through the data fitting model without requiring manual intervention. The model self-adjusts parameters by fitting user behavior data, eliminating the need for manual tag weight setting while maintaining adaptability through automated parameter optimization
Solution Approach 2:
The patent transforms the manual parameter setting process into an automated parameter optimization process. By using the data fitting model to automatically adjust tag weights based on user behavior data, the system changes from static manual configuration to dynamic automated parameter determination, resolving the contradiction between flexibility and time consumption
2Adaptability or versatility
If manual tag weight setting is used, then customization of user portrait parameters is improved, but accuracy of user portrait decreases due to potential errors
Solution Approach 1:
The data fitting model incorporates feedback mechanisms that automatically adjust tag weights based on user behavior data. The model continuously optimizes parameters by comparing predicted user preferences with actual behavior patterns, thereby improving accuracy while maintaining customization capabilities through data-driven parameter adjustment
Solution Approach 2:
The patent replaces the manual mechanical process of tag weight setting with an automated computational system. The data fitting model uses mathematical optimization algorithms to determine tag weights objectively, eliminating human errors associated with manual setting while preserving customization through flexible parameter adjustment based on user behavior data
3Productivity
If automated data fitting model is used, then time and labor costs are reduced, but system complexity increases
Solution Approach 1:
The data fitting model serves multiple functions simultaneously: it determines tag weights, optimizes user portrait parameters, and adapts to different user behavior patterns. By consolidating these functions into a single automated system, the patent reduces overall system complexity while maintaining high productivity through multi-functional parameter optimization
Data Source
AI summary
User portrait obtaining method, apparatus, and storage medium are provided. The method includes: obtaining M training samples according to a user behavior log. An initialized user parameter matrix Wm×k and an initialized tag parameter matrix Hk×n are modified according to the M training samples by using a data fitting model, to obtain a final user parameter matrix Wm×k and a final tag parameter matrix Hk×n. Further, a user portrait matrix Pm×n is obtained according to the final user parameter matrix Wm×k and the final tag parameter matrix Hk×n. In the present disclosure, a user and a tag are parameterized, and a user parameter matrix and a tag parameter matrix are modified by using the data fitting model, so as to fit a training sample.