Information Distribution Clustering Mobility Patterns
Solution Overview
Problem
Existing information distribution systems face limitations in efficiently distributing service information to a large number of users due to the requirement for substantial information processing and network resources, which can lead to constraints in the number of users that can be targeted effectively.
Innovation Solution
The system clusters users based on their mobility patterns and attributes, allowing for targeted information distribution within each cluster using optimized resources, thereby reducing the overall demand for processing power and network capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If service information is distributed to each user individually based on their mobility patterns and attributes, then the relevance and effectiveness of distributed information is improved, but the information processing resources and network resources required increase substantially
Solution Approach 1:
The patent merges multiple users with similar mobility patterns and attributes into clusters, treating them as a single unit for information distribution. This combining approach maintains information relevance by targeting users with similar characteristics while substantially reducing the computational resources needed compared to processing each user individually.
Solution Approach 2:
The patent creates representative mobility patterns that copy and generalize the behavior of multiple users within a cluster. Instead of processing each user's unique pattern, the system uses these representative copies to infer future positions and distribute information to entire clusters, reducing processing complexity while maintaining effectiveness.
2Productivity
If service information is distributed to a large number of users simultaneously, then the coverage and impact of information distribution is improved, but the network resources and processing capacity required increase, limiting the number of users that can be served
Solution Approach 1:
The patent combines multiple users into mobility pattern clusters to reduce the overall number of processing units. By distributing information to clusters rather than individual users, the system achieves broad coverage while consuming fewer network resources and processing capacity, thereby improving distribution efficiency.
Solution Approach 2:
The patent segments the user population into distinct mobility pattern clusters, allowing the system to manage and distribute information to large numbers of users in manageable groups. This segmentation enables scalable information distribution without requiring resources proportional to the total user count.
3Measurement precision
If detailed user attributes and past position information are processed for each user to estimate future positions, then the accuracy of position estimation is improved, but the computational time and processing power required increase
Solution Approach 1:
The patent merges the processing of detailed user attributes and position information across multiple users by clustering them together. This allows the system to maintain accurate position estimation based on detailed characteristics while reducing computational time through shared processing of common patterns among clustered users.
Solution Approach 2:
The patent performs preliminary clustering of users based on their mobility patterns and attributes before conducting detailed position estimation. This preliminary organization groups users with similar characteristics, enabling subsequent processing to leverage shared patterns and reduce redundant computations, thereby maintaining accuracy while decreasing processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Proposed are an information distribution apparatus and method capable of efficiently distributing information. Clustering processing of collecting time series position data, which is configured from position information of a plurality of mobile communication objects, and classifying the time series position data into a plurality of clusters based on the collected time series position data of each of the mobile communication objects is executed, profiling processing of determining, for each of the clusters, a dominant attribute of each user in which the time series position data was allocated to the relevant cluster based on pre-registered attribute information of each of the users of each of the mobile communication objects is executed, and, for each of the clusters, information to be distributed to each of the users in which the time series position data was allocated to the relevant cluster is determined based on a processing result of the profiling processing.