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

KNN classification-based optimization method

A technology of optimization method and classification method, which is applied in the application field of Internet data classification, can solve the problems affecting the classification results, achieve the effect of wide application range and improve the accuracy rate

Inactive Publication Date: 2017-05-31
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, when the traditional KNN classification method has unbalanced samples, for example, the sample size of one class is large, while the sample size of other classes is small, it may cause that when a new sample is input, the samples of the large-capacity class in the K neighbors of the sample accounted for the majority, thus affecting the classification results

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
  • KNN classification-based optimization method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0029] The main purpose of the present invention is to improve the KNN classification method, make up for the deficiency of the original KNN classification method, and improve the classification accuracy. The original KNN classification method mainly uses the similarity measure, and the classification basis is the maximum similarity of each category in the first K samples, that is,

[0030] (5)

[0031] (6)

[0032] in, representation with unknown samples nearest neighbor samples; Indicates the first kind; is the total number of categories;

[0033] The KNN original classification process is as follows:

[0034] Calculate the similarity between the data to be classified and each class, and select appropriate parameters , using formulas (5) and (6) to classify unknown samples.

[0035] When the KNN original classification method has unbalanced samples, such as the sample size of one class is large, while the sample size of other classes is small, it may ...

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

The invention discloses a KNN classification-based optimization method, and relates to the technical field of data classification methods. The method comprises the steps of introducing a bias parameter b for each category to adjust maximum similarity based on a conventional KNN classification method; and correcting the parameter b from a training sample through a learning algorithm. According to the method, an additional bias parameter is added for each category, so that the purpose of improving classification accuracy can be achieved.

Description

technical field [0001] The invention relates to the technical field of data classification methods, is suitable for the application of Internet data classification, and can improve the accuracy of information retrieval. Background technique [0002] With the development of information technology, the information that people can obtain has shown explosive growth. Facing the ever-increasing mass of information, it is becoming more and more difficult to process the information only by manual methods. Some automated auxiliary tools are needed to help people better manage and filter the information. K-Nearest Neighbor (KNN, K-Nearest-Neighbor) classification algorithm is one of the most commonly used methods in data mining classification technology. [0003] The so-called K nearest neighbors are the K nearest neighbors. The idea of ​​the KNN method is: samples belonging to the same category have similar characteristics, and the distribution in the feature space is uniform. There...

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): G06K9/62
CPCG06F18/241
Inventor 刘川汪文勇苟玲
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA