Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Definition and method of approximate local attribute reduction in decision table

A technology of local attributes and decision-making attributes, applied in knowledge expression, special data processing applications, instruments, etc., can solve problems such as high computational complexity, achieve the effect of reducing computational complexity and improving computational efficiency

Inactive Publication Date: 2017-12-12
BEIJING LANGUAGE AND CULTURE UNIVERSITY
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] The technical problem to be solved by the present invention is to provide a definition and method of approximately invariant local attribute reduction under the decision table to solve the problem of high computational complexity existing in the prior art

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
  • Definition and method of approximate local attribute reduction in decision table
  • Definition and method of approximate local attribute reduction in decision table
  • Definition and method of approximate local attribute reduction in decision table

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0038] The reduced definition of approximately invariant local attributes under the decision table provided by the embodiment of the present invention includes:

[0039] Obtain the decision table (U, C∪D), where U represents the domain of discourse, C represents the condition attribute set, D represents the decision attribute set, and both C and D are sets of equivalence relations on U;

[0040] Determine the quotient set of domain U about decision attribute set D as U / D={D 1 ,D 2 ,...,D k}, where D l Indicates the lth decision class of the decision table, the value of l is l∈{1,2,...,k}, and k is a positive integer;

[0041] Let set B be a non-empty subset of set C: For a given positive integer l∈{1,2,...,k}, if B satisfies:

[0042] (1) R C (D l ) = R B (D l ),in,R C (D l ) means to keep the decision class D l The equivalence relation R C The set of corresponding elements x of the lower approximation invariant, R B (D l ) means to keep the decision class D...

Embodiment 2

[0052] Such as figure 1 As shown, the embodiment of the present invention also provides a local attribute reduction method determined according to the definition of approximately constant local attribute reduction under the decision table, including:

[0053] S101, based on the attribute reduction algorithm that keeps the decision table approximately unchanged, calculate the D of the decision table l The resolution matrix of the decision class, where D l is a decision class for local attribute reduction;

[0054] S102. According to the obtained resolution matrix, convert the corresponding resolution function from the principal conjunctive normal form to the principal disjunctive normal form, and obtain all the reduction results of the local attribute reduction.

[0055] The method for reducing approximately invariant local attributes under the decision table described in the embodiment of the present invention defines the concept of approximately invariant local attribute re...

Embodiment 3

[0065] Such as figure 2 As shown, the approximate invariant local attribute reduction method under the decision table provided by the embodiment of the present invention will be described in detail. The local attribute reduction method includes:

[0066] S11, determine the decision table;

[0067] S12. According to the determined decision table, designate a certain decision class of the decision table used for local attribute reduction;

[0068] S13. Based on an attribute reduction algorithm that keeps approximately constant under the decision table, calculate the resolution matrix of the specified decision class;

[0069] S14. According to the obtained resolution matrix, the corresponding resolution function is converted from the principal conjunctive normal form to the principal disjunctive normal form, and all the reduction results of the local attribute reduction are obtained.

[0070] The method for reducing approximately invariant local attributes under the decision t...

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 present invention provides a definition and method of approximate local attribute reduction in a decision table. The definition and method make it possible to perform attribute reduction on a certain decision class of a decision table and reduce calculation complexity. The definition comprises: acquiring a decision table (U, C union of D), wherein, U represents a domain of discourse, C represents a set of condition attributes, and D represents a set of decision attributes; determining a quotient set of the domain of discourse U related to the set of decision attributes to obtain a decision class D1; and setting a set B as a non-empty subset of a set C, and if B meets a preset condition, calling B as an approximate local attribute reduction of C under the decision class D1. The method comprises: based on an attribute reduction algorithm that maintains an approximately invariant attribute in a decision table, calculating a resolution matrix of a D1 decision class of the decision table; and according to the obtained resolution matrix, converting a corresponding resolution function from a main coordination paradigm to a main abstraction paradigm, to obtain all reduction results of local attribute reduction. The definition and method provided by the present invention are applicable to attribute reduction of decision tables in a rough set.

Description

technical field [0001] The invention relates to the technical fields of data mining, knowledge discovery, pattern recognition and machine learning, in particular to a definition and method for approximately invariant local attribute reduction under a decision table. Background technique [0002] Attribute reduction is also called feature selection, which comes from machine learning. Attribute reduction has important applications in many fields, such as auxiliary decision-making, data mining, pattern recognition and other fields. [0003] In 2016, scholars such as X. Jia summarized 22 attribute reduction types, including positive area reduction, distribution reduction, variable precision reduction, coverage reduction, mutual information reduction, and cost-sensitive reduction. In fact, there are many more types of reduction. But at present, the research on the problem of attribute reduction basically belongs to the category of overall reduction of decision attribute values....

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): G06F17/30G06N5/02
CPCG06F16/2465G06N5/02
Inventor 李吉梅刘贵龙花正冯艳宾邹继阳
Owner BEIJING LANGUAGE AND CULTURE UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com