User Analysis System for Accurate Task Time Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating task completion time in video games are inaccurate for individual players due to broad distributions of completion times, as they rely on mean averages that fail to account for player proficiency and focus levels, leading to poor prediction for specific users.
Innovation Solution
A method and apparatus that analyze user data to estimate task completion time by characterizing player speed and focus, using sub-groups of users with similar completion times and paths, and selecting the most common or focused paths to provide more accurate predictions, incorporating historical data and current task states to refine estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If mean average completion times from a broad user corpus are used to estimate task completion time, then the estimation method is simple to implement, but the prediction accuracy for individual players is poor
Solution Approach 1:
The patent segments the broad user corpus into multiple sub-groups based on player characteristics such as completion speed and focus level. Instead of using a single mean average from all users, the system divides users into distinct segments (e.g., fast completers, focused players, distracted players) and selects the most appropriate sub-group whose mean completion time best matches the current player's profile, thereby improving prediction accuracy while maintaining computational feasibility.
2Measurement precision
If player data is segmented into sub-groups based on completion speed and focus, then prediction accuracy for individual players improves, but the complexity of the estimation system increases
Solution Approach 1:
The system segments players into sub-groups based on measurable characteristics like completion speed and focus level, allowing more accurate predictions without requiring complex individualized models for each player.
Solution Approach 2:
The patent changes the parameter of estimation from a single global mean to multiple sub-group means, where each sub-group is defined by specific parameters (completion speed, focus level). This allows the system to adapt the estimation parameter based on the player's characteristics, improving accuracy while keeping the underlying methodology consistent and manageable.
3Measurement precision
If the system uses completion routes and paths to characterize player behavior, then the estimation becomes more tailored to individual play styles, but the data processing requirements increase
Solution Approach 1:
The patent extracts specific relevant features from player data, such as completion routes and paths taken through the game, and uses these extracted characteristics to characterize player behavior and select appropriate sub-groups for estimation, rather than processing all available player data.
Solution Approach 2:
The system performs preliminary analysis of player completion routes and paths to establish player profiles and assign players to sub-groups before actual task estimation is needed. This preliminary characterization allows the system to quickly select appropriate estimation parameters without performing complex data processing during the estimation itself.
Data Source
AI summary
A method of estimating a time to complete a task for a user of a computer comprises: identifying a target task for which a time is to be estimated for completing the task, determining as a start point the completion of the last task completed by the user on the computer, obtaining, from data relating to a corpus of users, data for at least a subset of those users who completed the last task completed by the user before also completing the target task wherein the obtained data comprises timing data, calculating a representative time from the start point to the completion of the target task, based upon the obtained data, and implementing a predetermined action responsive to the calculated representative time.