Inference Apparatus for Tabular Data Labeling via Text Conversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for automatic labeling, such as zero-shot classification, are limited to text data and cannot handle data in tabular or other forms.
Innovation Solution
An inference apparatus that converts target data into text using data conversion rules and employs a label inference model to assign labels, allowing for the application of text-based label inference approaches to various data forms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If zero-shot classification is used for automatic labeling, then text data can be labeled automatically, but data in tabular form cannot be labeled
Solution Approach 1:
The patent introduces a data conversion unit as an intermediary component that transforms tabular data into text format. This mediator enables the text-based label inference model to process tabular data by converting it into a compatible format, thereby extending the model's applicability without requiring separate processing mechanisms for different data types.
Solution Approach 2:
The label inference model is designed with universal functionality to handle multiple data formats. By integrating the data conversion capability and using a unified text-based inference approach, the system achieves multi-functionality, allowing the same model to process both text and tabular data through a common processing pipeline.
2Reliability
If separate label inference models are created for different data forms, then each data type can be handled specifically, but the system becomes more complex
Solution Approach 1:
The patent employs a single unified label inference model that can handle multiple data formats through data conversion. This universal approach maintains reliability by using consistent inference logic across different data types while reducing system complexity compared to maintaining separate specialized models for each data format.
Solution Approach 2:
The system changes the parameter of data representation by converting tabular data into text format with specific structural characteristics. This parameter transformation allows the text-based inference model to process previously incompatible data types without modifying the model's core inference mechanism, thereby maintaining accuracy while simplifying the overall system architecture.
Data Source
AI summary
In order to apply a text-based label inference approach to infer a label to be assigned to target data in a common data form, an inference apparatus (1) includes: a data converting section (11) for converting target data subject to label assignment into text; and a label inferring section (12) for inferring a label to be assigned to the target data, in accordance with a label inference model for inferring a label to be assigned to text and the text obtained by the converting carried out by the data converting means.