Action Estimation System Using Degree-of-Interest Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
It is difficult to determine the actions a user is interested in, as existing systems lack the capability to estimate user interest effectively.
Innovation Solution
An action estimation system that obtains schedule and result information, generates degree-of-interest information for users, and estimates user interest by comparing this information with data from other users using a machine learning model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If schedule information and result information are collected and processed to generate degree-of-interest information, then user interest estimation accuracy is improved, but system complexity increases
Solution Approach 1:
The system divides the information processing into distinct segments: data collection (schedule information, result information), degree-of-interest calculation, and action estimation. This segmentation allows each component to be optimized independently while working together to achieve accurate user interest estimation without overwhelming complexity.
Solution Approach 2:
The degree-of-interest information acts as an intermediary between the raw data (schedule and result information) and the final action estimation. This intermediary layer simplifies the relationship between data collection and estimation, making the system more manageable while improving estimation accuracy through structured processing.
2Measurement precision
If degree-of-interest information is generated for multiple users and compared, then action estimation accuracy is improved, but information processing time increases
Solution Approach 1:
The system pre-calculates and stores degree-of-interest information for multiple users in advance. This preliminary processing allows the actual action estimation to be performed more quickly by simply comparing pre-computed data, rather than calculating everything in real-time, thus reducing processing time while maintaining accuracy.
Solution Approach 2:
The system creates and compares degree-of-interest information profiles for multiple users. By copying and comparing these profiles rather than analyzing each user's data independently in real-time, the system achieves accurate action estimation more efficiently, reducing the time required for information processing.
3Measurement precision
If the system processes schedule information and result information to estimate user interest, then recommendation accuracy is improved, but data storage requirements increase
Solution Approach 1:
The system extracts only the essential information needed for interest estimation from the raw schedule and result data. By taking out and storing only the degree-of-interest information rather than all原始 data, the system maintains recommendation accuracy while significantly reducing data storage requirements.
Solution Approach 2:
The system discards unnecessary raw data details after extracting the essential degree-of-interest information. This allows the system to retain only the necessary information for accurate recommendations while freeing up storage space, effectively balancing recommendation accuracy with storage efficiency.
Data Source
AI summary
An action estimation system includes: an obtainer that obtains at least one of schedule information or result information, the schedule information indicating a schedule of an action planned by a first user, the schedule of the action being for the first user in a predetermined period of time, the result information indicating an actual action of the first user in the predetermined period of time; a generator that generates first degree-of-interest information indicating a degree-of-interest of the first user for each of a plurality of actions based on the at least one of schedule information or result information obtained; and an estimator that estimates a first action that the first user is interested in other than the plurality of actions by comparing the first degree-of-interest information generated with each of a plurality of items of second degree-of-interest information each of which is generated for an other user.