Napovedovanje sprememb kognitivnih sposobnosti pri starejši populaciji

Štante, Anja

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Napovedovanje sprememb kognitivnih sposobnosti pri starejši populaciji
Authors:	ID Štante, Anja (Author) ID Štiglic, Gregor (Mentor) More about this mentor... ID Cilar Budler, Leona (Comentor)
Files:	MAG_Stante_Anja_2021.pdf (670,62 KB) MD5: 8D848E8A7F0EE4FC294B89B44B16216B PID: 20.500.12556/dkum/79521473-d872-4487-84f2-1be44a3df720
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FZV - Faculty of Health Sciences
Abstract:	Uvod: Delež starejših od 65 let se v populaciji veča. Napovedi kažejo, da se bo do leta 2100 še povečal. Ena od problematik, ki se pojavlja pri starejših, je upad kognitivnih sposobnosti. Metode: Analizirali smo podatke iz raziskave The Survey of Health, Aging and Retirement in Europe pri valu 6 in 7. Oblikovali smo dve skupini podatkov. V prvo smo uvrstili vse razpoložljive podatke iz držav Evropske unije, v drugo podatke zbrane v Republiki Sloveniji. Analizo podatkov smo izvedli v programskem jeziku R. Z uporabo napovednih modelov smo poskušali napovedati spremembe v vrednosti spremenljivke »razlika v takojšnjem priklicu besed med valoma 6 in 7«. Uporabili smo tri napovedne modele, in sicer multiplo logistično regresijo, Random Forest in Gradient Boosting Machine. Pri dobljenih rezultatih smo primerjali vrednost površine pod krivuljo. Rezultati: Pri evropski populaciji se je kot najboljši napovedni model izkazala multipla logistična regresija z rezultatom pri vrednosti AUC 0,698 (95% IZ: 0,678-0,705). Pri slovenski populaciji se je kot najboljši izkazal Gradient Boosting Machine z rezultati pri AUC 0,698 (95% IZ: 0,663-0,760). Razprava in sklep: Izkazalo se je, da na izhodno spremenljivko vplivajo spremenljivke, ki so že zaznane v drugih študijah. Te so priklic besed pri valu 6, starost in stopnja izobrazbe. Pri evropski populaciji je opazen vpliv spremenljivke samoocena računalniških sposobnosti. Študije s področja kognitivnih sposobnosti pri starejših na območju Republike Slovenije so pomembne, saj se slovenska družba stara.
Keywords:	staranje, kognitivni upad, priklic besed, SHARE, napovedni model
Place of publishing:	Maribor
Publisher:	[A. Štante]
Year of publishing:	2021
PID:	20.500.12556/DKUM-79124
UDC:	159.95:159.922.63(043.2)
COBISS.SI-ID:	68315395
Publication date in DKUM:	29.06.2021
Views:	1377
Downloads:	176
Metadata:
Categories:	FZV
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
Share:

Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:	CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International

Link:	http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:	The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:	10.05.2021

Secondary language

Language:	English
Title:	Predicting changes of cognitive abilities among older population
Abstract:	Introduction: In the population the share of people over 65 years old is increasing. Forecasts suggest that it will increase further by 2100. One of the problems that occurs among the elderly is a decline in cognitive abilities. Methods: We analyzed data from The Survey of Health, Aging and Retirement in Europe through waves 6 and 7. Two data sets were analyzed. In the first one was included data from the European Union, and in the second was included only data from Slovenia. The data analysis was performed in the programming language R. Using predictive models, we tried to predict changes in the variable "difference in the immediate recall of words between waves 6 and 7". We used three prediction models: Gradient boosting machine, multiple logistic regression, and Random Forest. In the obtained results we compared the value of the area under the curve. Results: In the European population, multiple logistic regression proved to be the best predictive model with AUC 0.698 (95% CI: 0.678-0.705). In the Slovenian population Gradient boosting machine proved to be the best with AUC 0.698 (95% CI: 0.663-0,760). Discussion and conclusion: It turned out that the input variable has been influenced by variables already detected in other studies. These are word recall at wave 6, age, and level of education. In the European population the influence of the variable self-assessment of computer skills is noticeable. Due to aging of Slovenian society, studies in the field of cognitive abilities in the elderly are important.
Keywords:	aging, cognitive decline, recall, SHARE, predictive model

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back