Prediction Model Using Base-Point Data for Regional Flow Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for predicting data, such as the number or density of people at a certain point and time, face challenges when input data is unavailable for certain regions due to privacy concerns or collection difficulties, leading to inaccurate stream-of-people estimation.
Innovation Solution
An information processing device that classifies regions into base points and their peripheries, builds a prediction model using data from base points to predict stream-of-people data for the peripheries, utilizing a periphery model to enhance accuracy even with incomplete input data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is collected from all regions to improve prediction accuracy, then measurement precision improves, but ease of operation deteriorates due to privacy concerns and collection difficulties
Solution Approach 1:
The patent segments the service area into multiple regions and further divides each region into a base point and periphery areas. This segmentation allows the system to collect data only from base points rather than all regions, reducing collection burden while maintaining prediction accuracy through the periphery model that estimates peripheral areas based on base point data.
Solution Approach 2:
The patent introduces base points as intermediary locations that serve as data collection hubs. These base points act as mediators between the data collection system and the peripheral areas, enabling indirect estimation of peripheral data through the periphery model without requiring direct data collection from all peripheral locations.
2Measurement precision
If data is collected from all regions to improve prediction accuracy, then measurement precision improves, but device complexity increases due to multiple data collection requirements
Solution Approach 1:
The system segments the service area into regions with base points and peripheries, simplifying the data collection architecture. Instead of requiring data collection infrastructure in all areas, the segmentation allows centralized collection at base points only, reducing system complexity while maintaining comprehensive coverage through model-based estimation.
3Measurement precision
If data is collected from all regions to improve prediction accuracy, then measurement precision improves, but loss of information increases due to privacy concerns
Solution Approach 1:
The patent extracts only the essential data needed for prediction from base points, rather than collecting comprehensive data from all regions. This extraction approach obtains sufficient information for accurate prediction while minimizing data collection intrusions and associated privacy concerns.
Data Source
AI summary
According to an embodiment, an information processing device includes one or more processors. The one or more processors are configured to: build a prediction model for predicting prediction data for a second region included in a plurality of regions based on acquired data acquired for one or more first regions included in the plurality of regions; and predict the prediction data based on the acquired data by using the prediction model.