Information Processing Apparatus Clustering Causal Graphs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In marketing and other fields, identifying important causal relationships that lead to problem-solving is challenging due to the complexity of large numbers of causal relationships generated by existing techniques.
Innovation Solution
An information processing program and apparatus that classify causal graphs and data groups into clusters based on similarity, allowing for the identification of important causal relationships by aggregating similar combinations and highlighting conditions under which these relationships appear.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If existing techniques are used to generate causal relationships, then comprehensive causal analysis is achieved, but the complexity of identifying important causal relationships increases due to large numbers of generated relationships
Solution Approach 1:
The patent segments the large set of causal relationships by classifying them into multiple clusters based on similarity. Each cluster represents a group of related causal relationships, making it easier to identify important ones without being overwhelmed by the total volume. The segmentation is performed through multiple classification stages using different similarity metrics.
Solution Approach 2:
The patent merges similar causal relationships into clusters by combining them based on similarity metrics. This merging process consolidates redundant or related relationships, reducing the overall complexity while preserving the essential causal information needed for comprehensive analysis.
2Measurement precision
If multiple classification methods are applied to organize causal relationships, then identification accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent divides the classification process into multiple sequential stages, each using a different similarity metric. The first classification uses one similarity measure, followed by a second classification using another metric. This segmented approach improves identification accuracy by examining relationships from multiple perspectives without requiring all classifications to occur simultaneously.
Solution Approach 2:
The patent performs preliminary classification using one similarity metric before performing subsequent classification with another metric. This preliminary action organizes the causal relationships in advance, making the subsequent classification more efficient and accurate while reducing the overall processing complexity compared to performing all classifications at once.
3Ease of operation
If similar data groups and causal graphs are aggregated into clusters, then ease of identification is improved, but computational requirements increase
Solution Approach 1:
The patent segments the computational task of aggregating similar data groups and causal graphs into multiple classification stages. By dividing the aggregation process into sequential steps using different similarity metrics, the computational load is distributed over time rather than requiring intensive simultaneous processing, making the operation easier while managing computational requirements.
Solution Approach 2:
The patent merges similar causal relationships and data groups into clusters based on similarity metrics. This merging consolidates redundant computations and allows the system to work with cluster representatives rather than individual relationships, reducing overall computational requirements while improving ease of identification.
Data Source
AI summary
A recording medium stores a program for causing a computer to execute a process including: referring to a memory storing data constituted by combinations of features to extract data groups of which the combinations satisfy each condition; identifying relationships between the features included in the data groups; classifying the relationships into a first clusters, based on first similarity; classifying the data groups into second clusters, based on second similarity; classifying the data groups into third clusters so as to classify, into a same cluster, data groups that are in a same one first cluster obtained by classifying the relationships corresponding correspond to each data group and are in a same one second cluster obtained by classifying each data group; identifying first conditions for classifying the data groups classified into each cluster and the data groups classified into other clusters; and outputting the identified first conditions with a classification result.