Mood Forecasting With Time-Interval Behavior Interpretation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mood forecasting methods fail to account for time-series changes in user emotions and lack interpretability, resulting in inaccurate average mood forecasts and a lack of understanding of effective behaviors for improving psychological states.
Innovation Solution
A computer executes two training procedures for neural networks using behavior and mood time-series data, divided into intervals, to forecast future moods and present the underlying behaviors supporting these forecasts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If data is separated and statistically processed on a daily basis to forecast average mood, then the forecasting process is simplified, but time-series changes in user mood at regular intervals are lost
Solution Approach 1:
The patent segments mood data into multiple time intervals within a day (e.g., morning, afternoon, evening) rather than processing it as a single daily average. This segmentation preserves the time-series characteristics of mood changes while still maintaining a structured forecasting approach. The system forecasts mood for each specific time interval separately, capturing the dynamic nature of mood fluctuations throughout the day.
2Ease of manufacture
If average mood values are forecasted, then the forecasting model is easier to implement, but the change in mood during the day cannot be presented to users
Solution Approach 1:
The patent adds the time interval dimension to the forecasting output. Instead of providing a single average mood value for the day, the system generates forecasts for multiple time intervals (morning, afternoon, evening), effectively moving from a one-dimensional average to a multi-dimensional time-series forecast. This allows users to understand how mood evolves throughout the day while maintaining model tractability.
3Measurement precision
If future mood is forecasted without presenting underlying behaviors, then the forecasting accuracy is maintained, but interpretability is lost and users cannot understand effective behaviors
Solution Approach 1:
The patent implements a feedback mechanism that presents both the forecasted mood values and the underlying behavioral factors that influenced the forecast. The system provides explanations such as 'mood is expected to improve because exercise time increased' or 'mood may decrease due to reduced sleep duration'. This feedback loop maintains forecast accuracy while enhancing user understanding of the relationship between behaviors and mood changes.
Data Source
AI summary
The present invention can forecast a future-mood and to present the forecasting and a behavior that is the basis of the forecasting by causing a computer to execute: a first-training procedure for training a first-neural network in accordance with behavior time-series data and mood time-series data in each time interval, the first-neural network using behavior time-series data and mood time-series data in a first-time interval as input to output a forecasted-value of behavior time-series data in a time interval following the first-time interval; and a second-training procedure for training a second-neural network in accordance with behavior time-series and mood time-series data per time interval, the second-neural network using the behavior time-series data in the following time interval and the mood time-series data in the first-time interval as input to output a forecasted-value of mood time-series data in the following time interval.