prikaz prve stranice dokumenta Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka
No public access
scientific master's theses
Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2001. urn:nbn:hr:168:942109
Vrančić, Ivan
University of Zagreb
Faculty of Electrical Engineering and Computing
Department of Electronic Systems and Information Processing

Institutional repository: FER Repository

Cite this document

Vrančić, I. (2001). Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka (Scientific master's theses). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:942109

Vrančić, Ivan. "Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka." Scientific master's theses, University of Zagreb, Faculty of Electrical Engineering and Computing, 2001. https://urn.nsk.hr/urn:nbn:hr:168:942109

Vrančić, Ivan. "Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka." Scientific master's theses, University of Zagreb, Faculty of Electrical Engineering and Computing, 2001. https://urn.nsk.hr/urn:nbn:hr:168:942109

Vrančić, I. (2001). 'Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka', Scientific master's theses, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 25 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:942109

Vrančić I. Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka [Scientific master's theses]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2001 [cited 2024 November 25] Available at: https://urn.nsk.hr/urn:nbn:hr:168:942109

I. Vrančić, "Metode raspoznavanja uzoraka za analizu poslovno-financijskih podataka", Scientific master's theses, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2001. Available at: https://urn.nsk.hr/urn:nbn:hr:168:942109

Metadata
TitleMetode raspoznavanja uzoraka za analizu poslovno-financijskih podataka
Title (english)Pattern Recognition Methods for Business Financial Data Analysis
AuthorIvan Vrančić
MentorSven Lončarić (mentor)
Committee memberSven Lončarić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
(Department of Electronic Systems and Information Processing)
Zagreb
Defense date and country2001, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Data Processing
Universal decimal classification
(UDC)		004 - Computer science and technology. Computing. Data processing
Abstract
Postoji jedan cross-selling problem, specifičan financijskim institucijama, koji je u okviru ovog magistarskog rada rješavan analizom klijenata upotrebom metoda raspoznavanja uzoraka. U radu je detaljno razrađena zaokružena cjelina koja se sastoji od procesa prikupljanja podataka o klijentima, procesa transformacije istih s ciljem izdvajanja značajki, procesa analize značajki klijenata izradom prediktivnih modela metodama raspoznavanja uzoraka i na kraju procesa implementacije rezultata analize u poslovni proces. Prediktivni modeli su napravljeni upotrebom više tehnika triju metoda raspoznavanja uzoraka - regresijske metode, metode stabla odluke i metode umjetnih neuronskih mreža. Opisani su načini mjerenja kvalitete prediktivnih modela na temelju kojih su napravljene usporedbe. Opisano je i kako je reorganizacijom podataka i izdvajanjem dodatnih značajki klijenata poboljšana kvaliteta prediktivnih modela. Poslovni stručnjaci su vrlo pozitivno ocijenili rezultate analize. U trenutku završetka izrade rada, intenzivno se radilo na implementaciji analize u poslovni proces.
Abstract (english)
Basic problem this work is dealing with is customer analysis using pattern recognition methods to solve a cross-selling problem, specific to financial institutions. The work describes the whole process cycle that consists of data collection and transformation, features extraction, analysis using predictive model development, and implementation of analysis results into business process. Predictive models are made using many techniques of three pattern recognition methods - regression, decision tree and artificial neural network. Several quality measurement methods are described and used to compare obtained models. Results are even improved by data reorganisation and extraction of extra features. Business experts are very pleased with analysis results. In work completion time, the complete analysis process has been implemented in business process.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:942109
Study programmeTitle: Postgraduate master programme in electrical engineering
Study programme type: university
Study level: postgraduate
Academic / professional title: Magistar znanosti elektrotehnike (Magistar znanosti elektrotehnike)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Created on2019-04-24 08:05:43