Sequential Data Classification with Correlation-Aware Likelihood Ratios
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sequential probability ratio test (SPRT) methods assume independent and identically distributed elements in sequential data, leading to insufficient accuracy due to neglecting the relationship between data elements.
Innovation Solution
An information processing device and method that calculates indicators for each element considering multiple elements, integrates these indicators to determine an integrated indicator, and classifies sequential data based on this integrated indicator, accounting for the relationships between elements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If SPRT method is used to classify sequential data, then the classification process is simple and efficient, but the accuracy is insufficient because the method does not consider the relationship between data elements
Solution Approach 1:
The patent segments the sequential data into individual elements while maintaining their sequence relationships. Each element is processed independently to calculate indicators, but these indicators are then integrated considering the sequential relationships between elements. This segmentation allows for both computational efficiency and accurate modeling of dependencies.
Solution Approach 2:
The patent introduces a new dimension of analysis by calculating indicators for each element while considering multiple elements simultaneously. Instead of treating elements as independent (0-dimensional) or only sequentially (1-dimensional), the method incorporates multi-element relationships as an additional dimension, enhancing classification accuracy without excessive complexity increase.
2Measurement precision
If the SPRT method assumes independent and identically distributed elements, then the mathematical formula is simple, but the classification accuracy deteriorates when data elements have strong correlations
Solution Approach 1:
The patent makes the classification approach dynamic by adapting to the correlation structure of the data. Instead of assuming independence, the method dynamically calculates indicators that reflect the actual relationships between elements in the sequential data, allowing it to adapt to various correlation patterns while maintaining computational tractability.
Solution Approach 2:
The patent changes the parameters used in the classification formula to account for element relationships. By modifying how indicators are calculated and integrated to reflect correlations between elements, the method maintains simplicity while improving accuracy for correlated data sequences.
3Measurement precision
If indicators are calculated for each element considering multiple elements, then the relationship between elements is captured, but the calculation time increases
Solution Approach 1:
The patent performs preliminary calculations of indicators for each element before final integration. By pre-computing these indicators while considering multiple elements, the method prepares the data in advance for the final classification decision, reducing the time needed for the actual classification process while maintaining accuracy.
Solution Approach 2:
The patent maintains continuous calculation of indicators as elements are processed sequentially. Instead of batch processing all elements at once, the method continuously computes and integrates indicators in real-time, reducing total calculation time while capturing element relationships throughout the sequence.
Data Source
AI summary
Provided is an information processing device including: an acquisition unit that sequentially acquires a plurality of elements included in sequential data; a first calculation unit that calculates, for each of the plurality of elements, indicators each indicating which of a plurality of classes is appropriate for corresponding element to belong to, in consideration of two or more elements among the plurality of elements; a second calculation unit that calculates, by integrating the indicators of the plurality of elements, an integrated indicator indicating which of the plurality of classes is appropriate for the sequential data to belong to; and a classification unit that classifies the sequential data into one of the plurality of classes based on the integrated indicator.