Personalized Relevance Modeling for Candidate Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current online services lack an accurate and efficient method to determine the relevance of user attributes, leading to the generation of irrelevant content and excessive resource consumption.
Innovation Solution
A computer system uses a personalized model to prompt users to select candidates based on their preferences for specific attributes, generating confidence scores based on the level of connection and online interaction between candidates and viewers, and displays a subset of candidates likely to be selected, thereby validating attribute relevance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the computer system generates recommendations based on all user profile attributes, then the coverage of recommendations is comprehensive, but the relevance of recommendations deteriorates due to inclusion of irrelevant attributes
Solution Approach 1:
The patent applies local quality by differentiating between relevant and irrelevant attributes within the user profile. Instead of treating all attributes uniformly, the system identifies and weights specific attributes based on their relevance to the user's actual preferences and behaviors, thereby improving recommendation quality while maintaining comprehensive coverage.
Solution Approach 2:
The system changes the parameter of attribute relevance by dynamically determining which profile attributes are actually relevant to each user through validation mechanisms. This allows the system to adjust the weight and importance of different attributes based on empirical validation data, resolving the contradiction between comprehensive coverage and relevant filtering.
2Adaptability or versatility
If the computer system displays all candidate content to users for selection, then the completeness of candidate selection is high, but the loss of time and resources increases due to users navigating through irrelevant content
Solution Approach 1:
The system applies partial action by displaying only a subset of candidate content that is most likely to be relevant to the user, rather than showing all candidates. This is achieved through validation scores and relevance modeling that predict which candidates the user is most likely to select, thereby reducing navigation time while maintaining selection completeness.
Solution Approach 2:
The system performs preliminary action by pre-validating attribute relevance and pre-ranking candidates before presentation to the user. Through validation scores derived from user selections and behavioral data, the system prepares and orders candidate content in advance, so users encounter relevant options first without having to navigate through irrelevant content.
3Quantity of substance
If the computer system generates and displays irrelevant content, then the quantity of content available is high, but the loss of energy and resources increases due to processing and displaying unnecessary information
Solution Approach 1:
The system extracts and removes irrelevant content from the recommendation pipeline through validation mechanisms. By identifying attributes and candidates with low validation scores, the system extracts only the relevant portion of content for processing and display, thereby maintaining content quantity quality while reducing energy consumption associated with processing and transmitting irrelevant information.
4Device complexity
If the computer system uses traditional attribute-based recommendation methods, then the simplicity of the system is maintained, but the accuracy of determining attribute relevance deteriorates
Solution Approach 1:
The system implements feedback loops where user selections and interactions with recommended content are fed back into the validation model. This feedback mechanism allows the system to continuously learn and improve its understanding of attribute relevance without requiring complex manual configuration, thereby improving accuracy while keeping the system architecture relatively simple through automated learning.
Data Source
AI summary
Techniques for selecting candidates using a personalized model are disclosed herein. In some embodiments, a computer system, for each candidate of a plurality of candidates, generating a corresponding confidence score for a combination of the candidate, a particular viewer, and a particular attribute based on a scoring model, with the corresponding confidence score being configured to indicate a likelihood that the particular viewer will select the corresponding candidate as a preference with respect to the particular attribute. The computer system then selects a subset of the plurality of candidates based on the corresponding confidence scores of the candidates in the subset, and causes the subset of candidates to be displayed on a computing device of the viewer along with a prompting for the viewer to select one of the selected subset of candidates as the preference with respect to the particular attribute.