Razvoj inteligentne rešitve za vrednotenje vozil z uporabo strojnega učenja

magistrsko delo

David Žele (Author), David Podgorelec (Mentor)

Abstract

Eden izmed ključnih problemov hitrega razvoja inteligentnih rešitev je v tem, kako čim bolj sistematično, avtomatizirano in nadzorovano optimizirati razvojni proces. Željen je čim manjši vpliv hitro vpeljanih sprememb v podatkih ali programski kodi na samo končno rešitev in sledenje vpliva sprememb. Namen magistrskega dela je predvsem vpeljati inženirski pristop ML Ops, ki v veliki meri avtomatizira korake v razvoju programske rešitve, in ga preizkusiti na primeru razvoja inteligentne rešitve za vrednotenje vozil z uporabo strojnega učenja in podatkov s spletnih strani z rabljenimi vozili. Končni rezultati, da je razvojni proces po vpeljavi ML Ops bolj organiziran ter v veliki meri razbremeni sodelujoče na projektu, ne presenečajo, preseneča pa medsebojna kompatibilnost izbranih orodij, ki smo jih izbrali za naš projekt, ter sama pripravljenost na vpeljavo koncepta.

Keywords

inteligentna rešitev;vrednotenje vozil;strojno učenje;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2022
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[D. Žele]
UDC:	004.85:004.413043.2)
COBISS:	146410243
Views:	148
Downloads:	39
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Development of an intelligent solution for vehicle evaluation using machine learning
Secondary abstract:	One of the key problems of rapid development of intelligent solutions is the problem of how to optimize the development process as systematically, automatically and controlled as possible. It is desired to have as little impact as possible when changes are rapidly introduced in data or program code and to track the impact of changes. The purpose of the master's work is primarily to introduce the ML Ops engineering approach, which largely automates the steps in the development of a software solution, and to test it on the example of the development of an intelligent solution for the evaluation of vehicles using machine learning and data from websites with used vehicles. The results, that the development process after the implementation of ML Ops was managed in a more organized manner and to a large extent relieves the project participants, are not surprising, but the mutual compatibility of the selected tools that we have chosen for our project and the readiness to introduce the concept, are quite surprising.
Secondary keywords:	intelligent systems;vehicle evaluation;machine learning;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	1 spletni vir (1 datoteka PDF (XII, 68 f.))
ID:	16562678