Data Analysis System Generating Business Actions via Awareness Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data analysis technologies can only generate actions for improving business performance based on explanatory variables closely correlated with the response variable, limiting the identification of all factors affecting performance and preventing the generation of actions from new perspectives.
Innovation Solution
A data analysis system that identifies target data based on the distribution of relevant indicators, calculates awareness features likely to contribute to business evaluation improvements, and generates actions to improve performance, allowing for the consideration of factors not solely correlated with the response variable.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If explanatory variables closely correlated with the response variable are used to generate actions, then the actions are effective for improving business performance, but the actions cannot be generated from new perspectives and may miss other factors affecting performance
Solution Approach 1:
The patent segments the data analysis process into multiple stages: first identifying explanatory variables through correlation analysis, then separately identifying awareness features through distribution analysis. This segmentation allows the system to generate actions based on both traditional correlation-based factors and novel distribution-based insights, resolving the contradiction between reliability and versatility.
Solution Approach 2:
The patent introduces an intermediary concept of 'awareness features' that bridge the gap between traditional correlation-based analysis and novel perspective-based actions. These awareness features are derived from distribution characteristics and serve as a mediator to generate actions that are both effective and innovative.
2Measurement precision
If traditional correlation-based analysis is used to identify factors affecting performance, then the analysis is straightforward and reliable, but it cannot identify factors from new perspectives
Solution Approach 1:
The patent merges two distinct analytical approaches: correlation-based identification of explanatory variables and distribution-based identification of awareness features. By combining these methods, the system achieves both measurement precision through correlation analysis and breadth of identification through distribution analysis, resolving the contradiction between accuracy and versatility.
Solution Approach 2:
The patent adds another dimension to the analysis by moving beyond traditional correlation-based single-dimensional analysis to include distribution-based multi-dimensional analysis. This dimensional expansion allows the system to identify factors from both traditional and novel perspectives simultaneously.
Data Source
AI summary
A system, which is configured to generate an action for controlling a business to be carried out by a business operation system, the system is configured to: obtain business data including a plurality of attributes relating to the business from the business operation system; identify target data to be analyzed based on a distribution of a relevant indicator; analyze the target data, to thereby calculate an awareness feature, which is likely to contribute to improvement of the business evaluation indicator; generate an action for improving the business evaluation indicator based on the awareness feature.