Unlock instant, AI-driven research and patent intelligence for your innovation.

Information processing device, creation method, and creation program

An information processing device and information volume technology, applied in neural learning methods, biological neural network models, machine learning and other directions, can solve problems such as reduced learning costs, inability to reuse models, and poor detection performance, and achieve low learning costs. Effect

Pending Publication Date: 2021-12-31
NIPPON TELEGRAPH & TELEPHONE CORP
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Here, if the learned model is generated without dividing the data, the detection performance will be deteriorated, but the learning cost will be reduced, and the reuse of the model will become possible.
On the other hand, if certain indicators such as IP addresses are used to divide the data to generate a learned model, there is a trade-off relationship as follows: although the detection performance becomes better, the learning cost becomes worse, and the model cannot be reused

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Information processing device, creation method, and creation program
  • Information processing device, creation method, and creation program
  • Information processing device, creation method, and creation program

Examples

Experimental program
Comparison scheme
Effect test

no. 1 approach

[0021] In the following embodiments, the configuration of the information processing device 10 of the first embodiment, the flow of processing of the information processing device 10 will be described in order, and finally the effects of the first embodiment will be described.

[0022] [Structure of the first embodiment]

[0023] First, refer to figure 1 The structure of the detection system of the first embodiment will be described. figure 1 It is a figure which shows an example of the structure of the detection system of 1st Embodiment. Such as figure 1 As shown, the detection system 1 has an information processing device 10 , a gateway 20 , and a device 30 , and the gateway 20 is connected to an external network 40 .

[0024] The information processing device 10 acquires the data of the normal state and the data of the detection object related to the equipment 30, and performs the learning of the acquired data of the normal state and the abnormality detection of the acqu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

An information processing device (10) calculates the information amount of respective dividing methods in each label for a data set divided by a label that serves as a candidate for an index when dividing data. Meanwhile, the information processing device (10) divides data into a plurality of data sets on the basis of a dividing method of the highest information amount among the calculated information amounts and creates a prelearned model for each data set using the divided data sets, respectively.

Description

technical field [0001] The present invention relates to an information processing device, a generation method, and a generation program. Background technique [0002] Conventionally, as a method for performing abnormality detection using teacher-less learning, a method of learning a probability distribution from normal data to generate a model is known. Here, if the learned model is generated without dividing the data, the detection performance will be deteriorated, but the learning cost will be reduced and the model can be reused. On the other hand, if some indicators such as IP addresses are used to divide the data to generate a learned model, there is a trade-off relationship as follows: although the detection performance becomes better, the learning cost becomes worse, and the model cannot be reused. In addition, there is also a method of changing the granularity of division and performing detailed inspections to find an appropriate granularity of division that does not...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N20/00
CPCH04L63/1425G06N3/088G06N20/00G06F16/906G06N3/047G06N3/045G06N7/01
Inventor 山田真德
Owner NIPPON TELEGRAPH & TELEPHONE CORP