Field Correlation Method for Binary Category Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional field collation systems struggle to accurately group fields with category values consisting of only 0 and 1, such as gender and married/unmarried fields, due to identical features making it difficult to determine corresponding fields within a limited number of fields.
Innovation Solution
A method and system that pre-store field groups, calculate correlation values between definite and indefinite fields, and group indefinite fields into new groups based on correlation thresholds, allowing for precise grouping of category fields by utilizing established correspondence relations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If field collation is performed using only data features (character type, average value), then the process is simple and fast, but fields with category values of 0 and 1 cannot be accurately grouped
Solution Approach 1:
The patent introduces an intermediary approach by using correlation values between fields as a mediator to bridge the gap between data features and field correspondence. Instead of directly matching fields based solely on data features, the system calculates correlation values between fields and uses these correlations as intermediary information to determine correspondence, thereby improving accuracy for binary category fields without completely redesigning the system architecture.
Solution Approach 2:
The patent adds another dimension to the field collation process by introducing correlation values as a new attribute for field comparison. Rather than relying only on traditional data features (character type, average value), the system now operates in a multi-dimensional space that includes correlation information, enabling more accurate discrimination between fields with identical data features.
2Adaptability or versatility
If multiple fields with identical data features are created, then data flexibility is improved, but it becomes difficult to determine which fields correspond to each other
Solution Approach 1:
The patent implements a feedback mechanism where the system calculates correlation values between fields and uses this information to refine field correspondence determination. The correlation calculation provides feedback about the relationships between fields, allowing the system to accurately identify corresponding fields even when multiple fields have identical data features, thus maintaining flexibility while improving identification accuracy.
Solution Approach 2:
The patent changes the parameters used for field comparison by adding correlation values as a new comparison criterion. When fields have identical data features, the system uses correlation values as an additional parameter to distinguish between them and determine correspondence, thereby resolving the ambiguity without restricting data structure flexibility.
3Extent of automation
If automatic field grouping is implemented, then manual handling is reduced, but fields with identical features cannot be distinguished
Solution Approach 1:
The patent uses correlation values as an intermediary that enables automated field grouping to work correctly for binary category fields. The correlation calculation serves as an intermediary step that provides the additional information needed for accurate distinction, allowing automation to proceed without sacrificing precision for fields with identical data features.
Solution Approach 2:
The patent performs preliminary correlation calculation before final field grouping. By pre-calculating correlation values between fields, the system prepares additional identification information in advance, which then enables accurate automated grouping of fields with identical data features without requiring manual intervention.
Data Source
AI summary
A field pair as a combination of a definite field and an indefinite field is decided and a correlation value between the definite field and the indefinite field in each of the field pairs is calculated. Among the field pairs in which the correlation value is not smaller than a threshold value, indefinite fields having corresponding definite fields which belong to the same field group are made to be a new field group.