magistrsko delo
Abstract
V našem delu smo se ukvarjali z implementacijo primerjalnika cen. Predstavili smo tehnološki sklad za implementacijo primerjalnika cen z uporabo oblačne tehnologije in pripravili prototip. Za implementacijo prototipa smo izbrali ponudnika oblačnih storitev Azure, ki je bil najbolj primeren glede na našo analizo. Prototip smo ocenili z več različnimi testi in določili ali je primeren za ta primer uporabe. Za implementacijo primerjalnika cen smo pripravili dva modela strojnega učenja za primerjavo artiklov, enega z uporabo pristopov analize besedila in drugega z uporabo pristopov slikovne analize. Modela smo med seboj primerjali, ocenili in predstavili boljšega za ta primer uporabe. V delu smo ugotovili, da je model z uporabo analize besedila primernejši model za ta primer uporabe, saj je zmožen natančneje primerjati večje število artilov, kot pa model slikovne analize.
Keywords
oblačna arhitektura;analiza slik;analiza besedila;Kubernetes;Azure;pridobivanje slik;mikrostoritve;Docker;dogodkovni tok;oblačna arhitektura aplikacij;računalništvo in informatika;magisteriji;
Data
Language: |
Slovenian |
Year of publishing: |
2023 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UL FRI - Faculty of Computer and Information Science |
Publisher: |
[A. Janc] |
UDC: |
004(043.2) |
COBISS: |
170122755
|
Views: |
59 |
Downloads: |
9 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
Design and implementation of a technological stack for efficient comparison of large amount of items |
Secondary abstract: |
In our work we implemented a price comparison application. We present a technological stack for the implementation of the application using cloud technologies and present a prototype of the implementation. We used Azure cloud provider for the implementation of the prototype, as it was most appropriate according to our analysis. We evaluated the prototype based on multiple tests and decide, if it is right for this use case. We also presented two machine learning models for comparing products, one model uses text analysis approaches and the other uses image analysis. We compared the models, evaluated their performance and present the better model for our use case. In our work we found that the text analysis model performs better for our use case, since it is able to compare products faster and more accurately. |
Secondary keywords: |
cloud architecture;image analysis;Kubernetes;Azure;image retrieval;microservices;Docker;event stream;cloud-native application architecture;computer science;computer and information science;master's degree;Računalništvo;Univerzitetna in visokošolska dela; |
Type (COBISS): |
Master's thesis/paper |
Study programme: |
1000471 |
Embargo end date (OpenAIRE): |
1970-01-01 |
Thesis comment: |
Univ. v Ljubljani, Fak. za računalništvo in informatiko |
Pages: |
107 str. |
ID: |
19963958 |