Generiranje besedila s pomočjo umetne inteligence

diplomsko delo

Tadej Lipar (Author), Sašo Karakatič (Mentor)

Abstract

V diplomski nalogi smo pregledali delovanje ter učinkovitost generiranja besedila z umetno inteligenco na več načinov. Dandanes računalniki generirajo vedno več besedila za prikaz človeku in ta besedila morajo biti čim bolj podobna človeku. S tem se sprosti človeško delo, katero se lahko uporabi za bolj kompleksno delo. S pomočjo grafov smo ugotovili, da je v tem primeru bolje uporabiti Markov model kot pa nevronsko mrežo. Generiranje besedila z računalnikom se najpogosteje uporablja za generiranje člankov v novinarstvu.

Keywords

generiranje besedila;Markov model;nevronska mreža;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2019
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[T. Lipar]
UDC:	004.777(043.2)
COBISS:	22835734
Views:	460
Downloads:	59
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Generating text with the help of artificial intelligence
Secondary abstract:	In this thesis we went over the working and efficiency of generating text with artificial intelligence with different methods. Nowadays computer generate more and more text meant for humans and the text must look like it was written by humans. With this the need for human written text is reduced and they can focus on more complicated work. With the help of graphs, we shoved that in this example the text generated with a Markov model is better or rather more human like than the text generated with a neural network. Generating text with computers is commonly used in reporting to generate articles.
Secondary keywords:	generating text;Markov model;neural network;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	42 f.
ID:	11204870