当前位置:网站首页>How to calculate the information entropy and utility value of entropy method?
How to calculate the information entropy and utility value of entropy method?
2022-06-25 08:23:00 【spssau】
One 、 application
Entropy is a measure of uncertainty . The more information , The less uncertainty , The smaller the entropy ; The less information , The greater the uncertainty , The greater the entropy . Therefore, the information carried by entropy is used to calculate the weight , Combined with the variation degree of each index , Using the tool of information entropy , Calculate the weight of each index , Provide basis for multi index comprehensive evaluation .
Two 、 operation
SPSSAU operation
(1) Click on SPSSAU Comprehensive evaluation ‘ Entropy method ’ Button . Here's the picture

(2) Drag and drop the data and click start analysis

Nonnegative translation : Because the calculation process of entropy method requires that there should be no 0 Or negative numbers , Otherwise, the result cannot be calculated . therefore SPSSAU This function is provided in the entropy method .
3、 ... and 、 Data processing
In order to eliminate the influence of dimensionality , Data processing is required before analysis .
(1) Positive indicators :(X-Min)/(Max-Min)
(2) Reverse indicators :(Max-X)/(Max-Min)
(PS: There are many dimensionless methods ,SPSSAU They are all available . Common treatment methods include standardization 、 normalization 、 Interval, etc .)
Four 、 analysis
Put data into analysis box ,SPSSAU The system automatically generates analysis results , as follows :

Calculation formula
- Information entropy e
(1) Computation first j Item... Under this indicator i Proportion of sample values

(2) Calculate the information entropy of each index ( Column )

In style ,k=1/ln(n);
2. Information utility value d

3. Weight factor

5、 ... and 、 summary
Use the entropy method to measure X1 Wait a total 5 Item for weight calculation , As can be seen from the table above :X1, X2, X3, X4, X5 in total 5 term , Their weight values are 0.311, 0.185, 0.352, 0.112, 0.040. There are certain differences in the weight of each item , among X3 The maximum weight of this item is 0.352, as well as X5 The lowest weight of this item is 0.040.
6、 ... and 、 Analysis and suggestion
SPSSAU Analysis and suggestion

边栏推荐
- 不怕百战失利,就怕灰心丧气
- Establish open data set standards and enable AI engineering implementation
- Bluecmsv1.6- code audit
- Opencv daily function structure analysis and shape descriptor (8) Fitline function fitting line
- Biweekly investment and financial report: capital ambush Web3 infrastructure
- Luogu p5994 [pa2014]kuglarz (XOR thinking +mst)
- Overview of image super score: the past and present life of image super score in a single screen (with core code)
- 企业全面云化的时代——云数据库的未来
- Matlab code format one click beautification artifact
- Deep learning series 45: overview of image restoration
猜你喜欢
![[thesis study] vqmivc](/img/38/a97ac763a7d6e71d4c7340c7abb6e7.png)
[thesis study] vqmivc

What is SKU and SPU? What is the difference between SKU and SPU

TCP acceleration notes

How to do factor analysis? Why should data be standardized?
![[QT] QT 5 procedure: print documents](/img/76/2fce505c43f75360a8ff477aa2d31d.png)
[QT] QT 5 procedure: print documents

Electronics: Lesson 013 - Experiment 14: Wearable pulsed luminaries

2022年毕业生求职找工作青睐哪个行业?

Wechat applet opening customer service message function development

共话云原生数据库的未来

Apache CouchDB Code Execution Vulnerability (cve-2022-24706) batch POC
随机推荐
Luogu p2048 [noi2010] super Piano (rmq+ priority queue)
打新债安不安全 有风险吗
双周投融报:资本埋伏Web3基础设施
Allgero reports an error: program has encoded a problem and must exit The design will be saved as a . SAV file
初体验完全托管型图数据库 Amazon Neptune
[red flag Cup] Supplementary questions
Luogu p3313 [sdoi2014] travel (tree chain + edge weight transfer point weight)
Luogu p6822 [pa2012]tax (shortest circuit + edge change point)
Remove headers from some pages in a word document
Jdbc-dao layer implementation
Electronics: Lesson 013 - Experiment 14: Wearable pulsed luminaries
以科技赋能设计之美,vivo携手知名美院打造“产学研”计划
Opencv minimum filtering (not limited to images)
420 sequence traversal of binary tree 2 (429. sequence traversal of n-ary tree, 515. find the maximum value in each tree row, 116. fill in the next right node pointer of each node, 104. maximum depth
Luogu p2839 [national training team]middle (two points + chairman tree + interval merging)
C # set up FTP server and realize file uploading and downloading
Electronics: Lesson 009 - Experiment 7: study relays
[QT] QT 5 procedure: print documents
Stm32cubemx Learning (5) Input capture Experiment
Thread + thread problem record