Sequential Data Alignment for Automated Model Labeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The process of generating a determination model and determining labels for data sets is labor-intensive due to the need for expert involvement in preprocessing, which varies by data type.
Innovation Solution
An information processing apparatus that includes an aligner to align sequential data sets and a target data extractor to automate the preprocessing of data, allowing for the automatic generation of determination models and label determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If preprocessing is performed manually by experts to ensure data accuracy and model performance, then determination accuracy is improved, but the effort and time required increases significantly
Solution Approach 1:
The system performs preprocessing automatically without requiring expert intervention. The preprocessing unit autonomously normalizes training data and determination target data, aligns sequential data sets, and extracts target data based on learned patterns from the neural network, eliminating the need for manual expert preprocessing while maintaining high determination accuracy
Solution Approach 2:
The patent replaces manual expert preprocessing operations with an automated computational system. The neural network learns appropriate preprocessing parameters from training data, and the preprocessing unit applies these learned parameters automatically, substituting the mechanical process of manual expert work with an intelligent automated system
2Manufacturing precision
If expert involvement is required to determine appropriate preprocessing methods for different data types, then data processing quality is improved, but device complexity and operational difficulty increase
Solution Approach 1:
The neural network serves multiple functions: it learns from training data, determines appropriate preprocessing parameters, and guides the extraction of target data from determination target data sets. This single universal component handles what would otherwise require multiple specialized expert operations, simplifying the overall system while maintaining high data processing quality
Solution Approach 2:
The system automatically adjusts preprocessing parameters based on the characteristics of the input data. The neural network learns optimal normalization parameters, alignment transformations, and extraction criteria by training on labeled data, allowing the system to adapt to different data types without requiring expert configuration for each case
Data Source
AI summary
An information processing apparatus according to an embodiment includes an aligner that aligns, with reference to a reference data set that is a sequential data set, another sequential data set; and a target data extractor that extracts a portion of the another sequential data set corresponding to the reference data set as a target data set.