Logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis

A Fisher discrimination, enterprise customer technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of generalization performance degradation, time-consuming and labor-intensive, and unlabeled samples remaining.

Inactive Publication Date: 2019-10-18
NORTHEAST FORESTRY UNIVERSITY
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, FDA and KFDA are supervised dimensionality reduction algorithms, which require a large amount of label information in advance to obtain better generalization performance.
However, in real life, especially labeling massive high-dimensional customer data of logistics companies is a very time-consuming and l

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
  • Logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis
  • Logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis
  • Logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0061] A kind of logistics enterprise customer classification method based on semi-supervised nuclear Fisher discriminant analysis, described Fisher discriminant analysis algorithm comprises the following steps:

[0062] In order to deal with multimodal or outlier problems, the present invention adopts local FDA (Local FDA, LFDA) algorithm, and its objective function is defined as:

[0063]

[0064] Here, S lb , S lw ∈R d×d represent the local inter-class scatter matrix and the local intra-class scatter matrix respectively, which are defined as:

[0065]

[0066]

[0067] Here W lb , W lw is an n×n matrix, and

[0068]

[0069]

[0070] represented in class y i ∈ The number of labeled samples in {1, 2, ..., c}, A ij is based on x i and x j measure of similarity between them.

[0071] For all i, j, when A ij = 1, LFDA is equivalent to the traditional FDA. Therefore, LFDA is also considered a localized version of traditional FDA. from In the defin...

Embodiment 2

[0090] A method for classifying logistics enterprise customers based on semi-supervised nuclear Fisher discriminant analysis, the semi-supervised nuclear Fisher discriminant analysis algorithm includes the following steps:

[0091] When there is unlabeled sample information in the sample set, in order to reasonably use the unlabeled sample information to guide the learning of supervised algorithms, it is necessary to consider the spatial consistency assumption of the sample set. For the LFDA dimensionality reduction method, spatial consistency is to ensure that the samples in the global and local neighborhoods in the original high-dimensional space still maintain this domain relationship in the dimensionality reduction space. Introducing the above spatial consistency assumption as a regularization term into the objective function of local Fisher discriminant analysis, the general framework of semi-supervised Fisher discriminant analysis is defined as

[0092]

[0093] in

...

Embodiment 3

[0114] In order to verify the logistics enterprise customer classification method based on semi-supervised nuclear Fisher discriminant analysis, this experiment classifies the customers of a large domestic logistics enterprise. According to the commonly used classification standard of the logistics enterprise obtained from the on-the-spot investigation, the logistics enterprise customers are divided into four categories: platinum customers, gold customers, diamond customers and ordinary customers.

[0115] In order to verify the classification performance of the logistics enterprise customer classification method based on the semi-supervised kernel Fisher discriminant analysis, the present invention first standardizes the logistics enterprise customer data, then introduces the semi-supervised idea, and uses the semi-supervised KFDA algorithm to standardize the logistics enterprise Classify customer data. Experimental environment: Windows7 operating system, CPU: Inteli7, 3.4G p...

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 logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis. The logistics enterprise customer classification method is characterized by comprising the following steps: (1) determining customer classification indexes and classification conditions commonly used by logistics enterprises; (2) collecting logistics enterprise customer information according to the customer classification indexes determined in the step (1); (3) standardizing the data sample set in the step (2); (4) constructing a consistency hypothesis matrix for the normalized customer sample data matrix obtained in the step (3), and calculating local inter-class and intra-class Laplace matrixes; (5) calculating a regularization term Laplacian matrixby utilizing the consistency hypothesis matrix obtained in the step (4), integrating the regularization term Laplacian matrix into a Fisher discriminant analysis target function, and obtaining an optimal projection matrix by solving a minimized target function; (6) calculating the projection coordinates of the normalized customer sample in the step (2) on the projection matrix; and (7) classifyingthe projection coordinates by using a nearest neighbor algorithm to determine the customer category. The logistics enterprise customer classification method is applied to classification of logisticsenterprise customers.

Description

Technical field: [0001] The application of the present invention relates to the field of logistics enterprise customer management, in particular to a logistics enterprise customer classification method based on semi-supervised kernel Fisher discriminant analysis. Background technique: [0002] In recent years, with the rapid development of IT and communication technologies such as the Internet, the Internet of Things, cloud computing, and triple play, big data has begun to spread to all walks of life in society. The advent of the era of big data has brought about major changes in the operating environment of domestic logistics companies. At present, domestic logistics is facing a market environment that is fully open and fully competitive at home and abroad. In the competitive business era, logistics companies must possess more advantageous resources, have a large number of effective customers, provide the best customer service, improve the original customer experience, and ...

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
IPC IPC(8): G06K9/62G06Q30/02
CPCG06Q30/0201G06F18/21322G06F18/21324G06F18/24147
Inventor 陶新民常瑞任超郭文杰李青刘锐陶思睿
Owner NORTHEAST FORESTRY UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap