Modeli grupiranja i klasifikacije krivulja mikciometrije

master's thesis

• Title (english): Clustering and Classification Models of Uroflow Curves
• Author: Josip Renić
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Marin Golub (član povjerenstva)
• Committee member: Marko Čupić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu promatrani su anonimizirani biološki vremenski nizovi krivulja mikciometrije snimanih kod djece. Provedena je dubinska analiza optimalnog broja grupa za dani skup podataka, analiza postupaka grupiranja i analiza klasifikacijskih postupaka s jasnim tumačenjem. Prikazan je opis razvijenog programa za izlučivanje značajki iz vremenskog niza mikciometrijske krivulje, također opisan je postupak izračuna značajki. Dobiven je optimalan broj od tri grupe, temeljen na internim mjerama kvalitete grupiranja. Naučeni su postupci klasifikacije s jasnim tumačenjem temeljeni na pravilima i stablom odluke. Cijeli postupak grupiranja i klasifikacije objedinjen je u web aplikaciji koja omogućuje učitavanje anonimiziranih zapisa vremenskih nizova mikciometrije te dobivanje odgovora o kategoriji obrasca krivulje, uz prikazani određeni postotak pouzdanosti kategorizacije.

Abstract (english)

In this paper anonymised biological time series of uroflow curves recorded in children were observed. A depth analysis of the optimum number of groups for a given data set, analysis of grouping procedures, and analysis of classification procedures with clear interpretation was performed. A description of the developed program for extracting features from the time series of the uroflow curve is also described, a feature calculation process is also described. An optimal number of three groups was obtained, based on internal grouping quality measures. The classification procedures with clear interpretation based on the rules and the decision tree have been learned. The entire grouping and classification process are combined into a web application that allows user to load anonymous uroflow curves and to obtain a response to the uroflow category, along with a certain percentage of categorization reliability.

• Keywords: Mikciometrija; grupiranje; klasifikacija; interne mjere kvalitete grupiranja
• Keywords (english): Uroflow; clustering; classification; internal clustering evaluation
• Language: croatian
• URN:NBN: urn:nbn:hr:168:412684
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 21:49:37