Analiza uspešnosti optične prepoznave elementov BPMN

magistrsko delo

Slavica Jagečić (Author), Gregor Polančič (Mentor)

Abstract

Magistrsko delo predstavlja postopek izdelave modela za prepoznavo ročno risanih BPMN elementov ter pridobitev rezultatov (%) uspešnosti njihove prepoznave. Za pomoč pri razvoju modela za prepoznavo elementov BPMN smo uporabili ogrodje TensorFlow. Opravili smo pregled literature, predstavili obstoječe rešitve, razvite na podlagi optične prepoznave in strojnega učenja. Razložili smo osnovne gradnike BPMN (standard BPMN 2.0.) in nekatere od teh elementov vključili v proces analize uspešnosti razpoznave s pomočjo mobilne aplikacije, izdelane v okviru naloge in razvite v okolju Angular.js, v katero smo vključili izdelani TensorFlow model, ki je zmožen prepoznavati BPMN elemente. V analizi smo zapisali ugotovitve, ki smo jih pridobili v raziskovalnemu delu na podlagi vprašalnikov. Ugotovitve, pridobljene v analizi, so pokazale da je mobilna aplikacija zmožna prepoznavati določene elemente BPMN, vendar ne vseh. Prav tako smo podali smernice za nadaljnje delo.

Keywords

analiza uspešnosti;izdelava modela;strojno učenje;mobilne aplikacije;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2019
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[S. Jagečić]
UDC:	004.424.3(043.2)
COBISS:	22839574
Views:	521
Downloads:	76
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Analysis of the effectiveness of optical recognition of BPMN elements
Secondary abstract:	The master thesis represents the process of creating a model capable of recognizing handwritten BPMN elements and receiving the results with percent precision (%) of recognition success. To create a model for identifying BPMN elements, we used the TensorFlow framework. We made a literature review, explored existing solutions for optical recognition and machine learning. We also explained the basic BPMN (BPMN 2.0. specification) elements and used some of those elements in the process of analyzing the recognition performance with the use of a mobile application, developed with Angular.js framework. In this application, we incorporated the TensorFlow model capable of detecting BPMN elements. In the analysis, we recorded the findings obtained in the research work based on a survey. The findings obtained in the analysis showed that the mobile application is capable of identifying certain BPMN elements, but not all. We also gave guidelines for further work.
Secondary keywords:	BPMN;OCR;machine learning;TensorFlow;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	XI, 96 str.
ID:	11210977