18267730. INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER PROGRAM simplified abstract (NEC Corporation)
INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER PROGRAM
Organization Name
Inventor(s)
INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER PROGRAM - A simplified explanation of the abstract
This abstract first appeared for US patent application 18267730 titled 'INFORMATION PROCESSING SYSTEM, INFORMATION PROCESSING METHOD, AND COMPUTER PROGRAM
Simplified Explanation
The patent application describes an information processing system that can classify series data into different classes based on a likelihood ratio calculated using elements of the data.
- Acquisition unit obtains multiple elements from series data.
- Calculation unit calculates a likelihood ratio indicating the likelihood of the series data belonging to a certain class, based on consecutive elements.
- Classification unit categorizes the series data into one of multiple classes based on the likelihood ratio.
- Learning unit improves the calculation of the likelihood ratio using a loss function that adjusts the ratio based on correct answer classes.
- Potential Applications
- Machine learning algorithms
- Pattern recognition systems
- Data classification tools
- Problems Solved
- Efficient classification of series data
- Improved accuracy in categorizing data
- Enhanced learning capabilities for information processing systems
- Benefits
- Increased accuracy in classifying data
- Faster processing of series data
- Enhanced machine learning capabilities
Original Abstract Submitted
An information processing system includes: an acquisition unit that obtains a plurality of elements included in series data; a calculation unit that calculates a likelihood ratio indicating a likelihood of a class to which the series data belong, on the basis of at least two consecutive elements of the plurality of elements; a classification unit that classifies the series data into at least one class of a plurality of classes that are classification candidates, on the basis of the likelihood ratio; and a learning unit that performs learning related to calculation of the likelihood ratio, by using a loss function in which the likelihood ratio increases when a correct answer class to which the series data belong is in a numerator of the likelihood ratio and the likelihood ratio decreases when the correct answer class is in a denominator of the likelihood ratio.