Klasifikacija studenata s obzirom na profesionalne i edukativne aktivnosti tijekom studija

master's thesis

• Title (english): CLASSIFICATION OF STUDENTS ACCORDING TO THEIR PROFESSIONAL AND EDUCATIONAL ACTIVITIES DURING COLLEGE
• Author: Vlatka Andrić
• Mentor: Mirjana Pejić Bach (mentor)
• Committee member: Ivan Strugar (predsjednik povjerenstva)
• Committee member: Jovana Zoroja (član povjerenstva)
• Granter: University of Zagreb; Faculty of Economics and Business; Zagreb
• Defense date and country: 2015-09-14, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Economics

Abstract

Ovaj rad bavi se problemom predviđanja profesionalnih i edukativnih aktivnosti studenata koji uz svoje fakultetske obaveze, rade preko student servisa i/ili se dodatno usavršavaju, odnosno uče strane jezike. Ovaj problem je zanimljiv iz razloga što govori o navikama mladih ljudi. Nakon što su utvrđeni faktori za koje se očekivalo da mogu utjecati na aktivnosti studenata, formirana je anketa te je provedeno istraživanje među studentima Menadžerske informatike na Ekonomskom fakultetu u Zagrebu. Uspješno je anketirano 120 ispitanika, odnosno 72 studenata i 48 studentica. U radu su prezentirani najvažniji faktori koji utječu na aktivnosti studenata te su isti korišteni i za izgradnju modela stabla odlučivanja. Osim modela stabla odlučivanja, u ovom radu su detaljno opisane i druge klasifikacijske metode, odnosno klasifikacija temeljena na pravilima, metoda k-najbližih susjeda i Bayesov klasifikator.
Iz ankete je utvrđeno 13 varijabli o kojima ovise aktivnosti studenata. Klasifikacijski modeli koji su sačinjeni od 7 najznačajnijih varijabli imaju prednost u odnosu na modele koji su izgrađeni od 13 varijabli. Razlog tome je što manji modeli imaju veću točnost, robusnost, jednostavniju interpretaciju i laku razumljivost.

Abstract (english)

In this work we are facing the problem of anticipating professional and educational activities of students who apart from their college obligations have extra work through student job center, and/or learn foreign languages. This problem is particularly important because it involves young people habits. After detecting factors for which we expected that they could affect student activities we have conduct a survey among the students of Managerial Informatics at the Faculty of Economics & Business Zagreb. The survey included 120 participants (72 male and 48 female students). In this work we have introduced the most important factors which affect student activities and these factors have also been used for making the Decision tree model. Apart from the Decision tree model we have also described other classification methods like Rule-based classification method, K-nearest neighbours and Bayesian classifer.
In this survey we have established 13 factors on which student activities depend. Classification models that are made of the most important 7 variables have significant advantage over the models made of 13 variables. The reason for this is because smaller models are more acuurate, they can easily be interpreted and they are more comprehensible.

• Keywords: otkrivanje znanja; rudarenje podataka; klasifikacija podataka; problem klasifikacije; metode klasifikacije; stabla odlučivanja; overfitting; klasifikacija temeljena na pravilima; metoda k najbližih susjeda; Bayesova mreža; ocjena klasifikacijskih modela; klasifikator; ROC krivulja
• Keywords (english): data mining; classification; decision trees; classification problems; classification methods; overfitting; rule-based classification method; k-nearest neighbours; Bayesian classifer; evaluation of classification methods; ROC curve
• Language: croatian
• URN:NBN: urn:nbn:hr:148:984413
• Study programme: Title: Business Economics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra ekonomije (magistar/magistra ekonomije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Access restricted to students and staff of home institution
• Created on: 2016-10-11 17:00:35