Implementacija i vrednovanje algoritma za izgradnju stabla odluke C4.5

undergraduate thesis

• Title (english): Implementation and evaluation of C4.5 decision tree algorithm
• Author: Luka Kiseljak
• Mentor: Alan Jović (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Nikolina Frid (član povjerenstva)
• Committee member: Ivan Kovačević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu iznesen je osnovni princip rada algoritma stabla odluke te su objašnjeni ključni koncepti koji se koriste pri izgradnji uspješnog stabla odluke na razini algorimta ID3. Na to je prezentiran i niz nadogradnji koji je implementiran u njegovom nasljedniku, algoritmu C4.5, koje uključuju rad s numeričkim atributima, nošenje s nepoznatim vrijednostima te postupke koji povećavaju općenitost stabla poput prestanka grananja prije nego su svi atributi poznati i podrezivanje. Opisana je i testirana ponuđena implementacija na dva skupa podataka koji su dali dobar uvid u sposobnosti implementiranog rješenja.

Abstract (english)

Contains basic principle behind decision tree algorithm as well as an introduction to key concepts used when building a successful decision tree on level of ID3 algorithm. On top of that a series of enhancements are presented that are implemented in its succesor, C4.5 algorithm. These include working with numerical values, handling unknown data and various procedures used to generalize trees, such as pruning and stoping tree development earlier than when all attributes are used up. A description and a comperhensive test of the developed implementation is also included giving a good insight in the capabilites of the provided solution.

• Keywords: Umjetna inteligencija; problem klasifikacije; strojno učenje; stabla odluke; ID3; C4.5; vrednovanje klasifikatora; točnost; preciznost; osjetljivost; F1 mjera
• Keywords (english): Artificial intelligence; classification problem; machine learning; decision tree; ID3; C4.5; classifier evaluation; accuracy; precision; recall; F1 measure
• Language: croatian
• URN:NBN: urn:nbn:hr:168:854103
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 21:32:05