User Score Prediction via Multi-Dimensional Pairwise Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing AI-based user score prediction methods require large amounts of training data and are costly, with inferior accuracy for predicting test scores due to insufficient actual test score data, especially in online education settings where data collection is limited.
Innovation Solution
A user score prediction method and system that uses multi-dimensional pairwise comparison of response information between users to predict scores, reducing data requirements and increasing prediction speed by training AI models on question solving information from multiple users, allowing for high-accuracy score predictions without collecting individual test scores.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing AI-based prediction methods use large amounts of training data including individual question solving data and test scores, then prediction accuracy can be improved, but data collection cost and processing time increase significantly
Solution Approach 1:
The patent extracts only the essential comparison information needed for prediction - specifically whether user A's score is higher, lower, or equal to user B's score - rather than using complete individual test score data. This extraction of core relational information reduces data requirements while maintaining prediction capability
Solution Approach 2:
The patent introduces response comparison information as an intermediary representation that mediates between raw individual responses and final score predictions. This intermediary form captures the essential scoring relationships without requiring complete individual test data, enabling accurate predictions with reduced data input
2Measurement precision
If individual test scores of users are collected one by one for training AI models, then accurate score prediction can be achieved, but data collection cost and time consumption increase
Solution Approach 1:
The patent extracts only the comparative relationship information (who scored higher, lower, or equal) from complete individual test scores. This extraction eliminates the need to collect and process entire test score datasets, reducing data collection time and cost while preserving the essential information needed for accurate score prediction
Solution Approach 2:
The patent uses partial information (comparison outcomes) rather than complete information (full test scores) to achieve the prediction goal. This partial action approach reduces data collection requirements while maintaining sufficient accuracy for practical applications
3Adaptability or versatility
If complete question solving data of all users is used for AI model training, then comprehensive prediction capability is achieved, but processing complexity and computational cost increase
Solution Approach 1:
The patent extracts essential comparative relationships from complete question solving data, transforming complex individual response patterns into simplified score comparison outcomes. This extraction reduces processing complexity while maintaining the adaptability needed for accurate score prediction across different users and contexts
Solution Approach 2:
The patent changes the data representation parameter from detailed individual question solving data to aggregated comparison outcomes (higher/lower/equal relationships). This parameter transformation simplifies the input data structure, reducing computational complexity while preserving the essential information for versatile prediction capability
Data Source
AI summary
Provided is a user score prediction device capable of reducing the amount of data used and increasing a prediction speed by predicting a score using only response comparison information obtained by mutually comparing responses of a plurality of users according to an embodiment of the present disclosure.The user score prediction device for predicting a score of a user through multi-dimensional pairwise comparison according to an embodiment of the present disclosure includes a question solving information collection unit that provides a question to be solved to a user through a user terminal and collects question solving information solved by the user, a response comparison information generation unit that generates response comparison information by performing multi-dimensional pairwise comparison between question solving data of a new user and question solving data of a reference user, and a score prediction unit that inputs the response comparison information to an artificial intelligence (AI) model trained in advance by response comparison information between arbitrary users and test scores to predict a score of the new user, and transmits the predicted score to a user terminal of the new user.