prikaz prve stranice dokumenta Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach
  Download
PDF 1.81 MB
undergraduate thesis
Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach
2019. urn:nbn:hr:137:259944
Simić, Srđan Daniel
University of Pula
Faculty of Informatics in Pula

Cite this document

Simić, S. D. (2019). Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach (Undergraduate thesis). Pula: University of Pula. Retrieved from https://urn.nsk.hr/urn:nbn:hr:137:259944

Simić, Srđan Daniel. "Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach." Undergraduate thesis, University of Pula, 2019. https://urn.nsk.hr/urn:nbn:hr:137:259944

Simić, Srđan Daniel. "Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach." Undergraduate thesis, University of Pula, 2019. https://urn.nsk.hr/urn:nbn:hr:137:259944

Simić, S. D. (2019). 'Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach', Undergraduate thesis, University of Pula, accessed 03 November 2024, https://urn.nsk.hr/urn:nbn:hr:137:259944

Simić SD. Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach [Undergraduate thesis]. Pula: University of Pula; 2019 [cited 2024 November 03] Available at: https://urn.nsk.hr/urn:nbn:hr:137:259944

S. D. Simić, "Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach", Undergraduate thesis, University of Pula, Pula, 2019. Available at: https://urn.nsk.hr/urn:nbn:hr:137:259944

Please login to the repository to save this object to your list.
Metadata
TitleAutomated Decision Modeling with DMN and BPMN: A Model Ensemble Approach
Title (english)Automated Decision Modeling with DMN and BPMN: A Model Ensemble Approach
AuthorSrđan Daniel Simić
MentorDarko Etinger (mentor)
MentorNikola Tanković (komentor)
Committee memberGiorgio Sinković (predsjednik povjerenstva)
Committee memberDarko Etinger (član povjerenstva)
Committee memberNikola Tanković (član povjerenstva)
GranterUniversity of Pula
(Faculty of Informatics in Pula)
Pula
Defense date and country2019-09-27, Croatia
Scientific / art field,
discipline and subdiscipline		SOCIAL SCIENCES
Information and Communication Sciences
Information Systems and Information Science
Abstract
Pregršt dostupnih heterogenih tranzakcijskih podataka te nepredak unutar područja strojnog učenja služe kao ključne sile za razvoj kompleksnih algoritama koji uspijevaju postići preformanse ljudske razine na sve većem broju zadataka. Zbog svoje ne linearne strukture sastavljene od mnogo slojeva kompjutacije, ti modeli visoke preciznosti obično su primjenjeni kao crne kutije; bez dubljeg razumjevanja o njihovim unutarnjim mehanizmima. To smanjuje transparentost procesa donošenja odluka i često može sadržati skrivene predrasude koji su potencijalno skriveni u podacima. Unutar ovog rada predlažemo alat za generiranje modela za donošenje odluka koji je u skladu s Decision Model & Notation (DMN) standardom baziran na tehnikama smanjivanja kompleksnosti. Ansambl klasifikatora stabla odlučivanja slojevite arhitekture je predložen kako bi kontrolirao "bias-variance trade-off". Evaluirali smo preformanse predložene metode na nekoliko javno dostupnih setova podataka usko vezanih za socijalno osjetljivo odlučivanje. Unutar ovog rada zanima nas automatizacija donošenja odluka koristeči DMN zajedno sa poznatim BPMN alatom za automatizaciju procesa. Primarnih ishod ovog istraživanja je okvir za modeliranje procesa odlučivanja baziranog na prijašnjim ishodima odluka primjenom tehnika strojnog učenja na način bijele kutije. Razumjevanje izlaza modela i zašto je do tog izlaza došlo ključno je za izbjegavanje slučajnih predrasuda. Naši doprinosi su sljedeći: (1) ansambl metoda strojnog učenja koja je semantički kompatibilna s DMN specifikacijama, time i razumljiva čovjeku, (2) okvir za primjenu ove nove metode na već postojećim zapisima odluka, (3) algoritam za automatsku pretvorbu predloženih modela u DMN artefakte.
Abstract (english)
Plethora of available heterogeneous transactional data and recent advancements in machine learning are the key forces that enable the development of complex algorithms that can reach human-level performance on an increasing number of tasks. Given the non-linear structure composed of many layers of computation, these highly accurate models are usually applied in a black-box manner: without a deeper understanding of their inner mechanisms. This hinders the transparency of the decision-making process and can often incorporate hidden decision biases that are potentially present in the data. We propose a framework for generating decision-making models conforming to Decision Model & Notation standard based on complexity-reducing techniques. An ensemble of decision-tree classifiers in a layered architecture is proposed to control the bias-variance trade-off. We have evaluated the performance of the proposed method on several publicly available data-sets tightly related to socially sensitive decision-making.
Keywords
Keywords (english)
Languagecroatian
Languageenglish
URN:NBNurn:nbn:hr:137:259944
Study programmeTitle: Informatics
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informatike (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informatike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2019-10-01 10:59:51