| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font
First page > Show document

Show document Help

Title:Metode za napovedovanje intervalnih spremenljivk na družboslovnih podatkih : diplomsko delo visokošolskega študijskega programa Informacijska varnost
Authors:ID Blagojević, Danijel (Author)
ID Mihelič, Anže (Mentor) More about this mentor... New window
Files:.pdf VS_Blagojevic_Danijel_2024.pdf (1006,06 KB)
MD5: 6FE4AFEF737D76F7C13860D49D5E1063
 
Language:Slovenian
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FVV - Faculty of Criminal Justice and Security
Abstract:Diplomsko delo obravnava uporabo metod podatkovnega rudarjenja za napovedovanje intervalnih spremenljivk v družboslovnih podatkovnih množicah, kjer je analiza velikih in kompleksnih podatkov pogosto izziv. Podatkovno rudarjenje omogoča odkrivanje skritih vzorcev in povezav, ki jih z običajnimi metodami analize težko zaznamo, ter omogoča napovedovanje prihodnjih dogodkov na osnovi preteklih podatkov. V nalogi so preučene različne metode, kot so linearna regresija, nevronske mreže, metoda k najbližjih sosedov (kNN), podporni vektorji (SVM) in odločitvena drevesa, pri čemer se raziskuje njihova učinkovitost pri obdelavi družboslovnih podatkov. Naloga vključuje primerjavo učinkovitosti teh metod z uporabo različnih metrik, kot so MSE, RMSE, MAE in R2, ki zagotavljajo celovito oceno natančnosti napovedi. Rezultati kažejo, da so nevronske mreže in linearna regresija med najuspešnejšimi metodami za napovedovanje intervalnih spremenljivk v družboslovnih podatkovnih množicah. Nevronske mreže izkazujejo posebno prednost pri odkrivanju skritih vzorcev v kompleksnih podatkovnih strukturah, medtem ko je linearna regresija učinkovita zaradi svoje preprostosti in razumljivosti. Raziskava poudarja pomen prilagajanja izbranih metod značilnostim podatkovnih množic in potrebam analize. Izpostavlja tudi, da lahko ustrezna izbira metode vpliva na uspešnost napovedovanja in natančnost pridobljenih rezultatov. S tem prispeva k razumevanju optimalnih pristopov za analizo družboslovnih podatkov in odpira možnosti za nadaljnje raziskave, zlasti pri uporabi naprednejših tehnik strojnega učenja in prilagoditvah obstoječih metod. Naloga tako ponuja smernice za izbiro ustreznih metod podatkovnega rudarjenja glede na zahteve družboslovnih raziskav in spodbuja nadaljnje raziskovalno delo na tem področju.
Keywords:podatkovno rudarjenje, metode, družboslovje, množice, metrike, diplomske naloge
Place of publishing:Ljubljana
Place of performance:Ljubljana
Publisher:D. Blagojević
Year of publishing:2024
Year of performance:2024
Number of pages:IX f., [40] str.
PID:20.500.12556/DKUM-90477 New window
UDC:303.7(043.2)
COBISS.SI-ID:209281795 New window
Publication date in DKUM:27.09.2024
Views:0
Downloads:25
Metadata:XML DC-XML DC-RDF
Categories:FVV
:
BLAGOJEVIĆ, Danijel, 2024, Metode za napovedovanje intervalnih spremenljivk na družboslovnih podatkih : diplomsko delo visokošolskega študijskega programa Informacijska varnost [online]. Bachelor’s thesis. Ljubljana : D. Blagojević. [Accessed 23 April 2025]. Retrieved from: https://dk.um.si/IzpisGradiva.php?lang=eng&id=90477
Copy citation
  
Average score:
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and 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:04.09.2024

Secondary language

Language:English
Title:Methods for predicting interval variables in social science data
Abstract:This thesis explores the use of data mining methods for predicting interval variables in social science data datasets, where the analysis of large and complex data is often challenging. Data mining enables the discovery of hidden patterns and connections that are difficult to detect using conventional analysis methods and allows for the prediction of future events based on past data. The thesis examines various methods, such as linear regression, neural networks, k-nearest neighbors (kNN), support vector machines (SVM), and decision trees, and investigates their effectiveness in processing social science data. The thesis includes a comparison of the effectiveness of these methods using various metrics, such as MSE, RMSE, MAE, and R2, which provide a comprehensive assessment of prediction accuracy. The results show that neural networks and linear regression are among the most successful methods for predicting interval variables in social science datasets. Neural networks demonstrate a particular advantage in uncovering hidden patterns in complex data structures, while linear regression is effective due to its simplicity and interpretability. The research emphasizes the importance of adapting the selected methods to the characteristics of data sets and the specific needs of the analysis. It also highlights that choosing the appropriate method can significantly impact the success of predictions and the accuracy of the obtained results. This contributes to understanding optimal approaches for analyzing social science data and opens opportunities for further research, especially in the use of advanced machine learning techniques and adaptations of existing methods. The thesis thus provides guidelines for selecting suitable data mining methods based on the requirements of social science research and encourages further exploration in this field.
Keywords:data mining, methods, social sciences, datasets, metrics


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica