diplomsko delo
Andraž Zrimšek (Author), Iztok Lebar Bajec (Mentor)

Abstract

Avtomatsko strojno prevajanje je lahko za podjetja zelo koristno, ker uporabnikom, ki ne govorijo angleščine, omogoča uporabo programske opreme in poslovanje v izbranem jeziku. Storitve strojnega prevajanja ponujajo številna podjetja, vendar te pri prevodih domensko specifičnih besedil ponavadi ne dosežejo dovolj visoke natančnosti. Za reševanje tega problema nekateri ponujajo storitve z možnostjo uglaševanja prevajalnikov s svojimi podatki. Med njih spada tudi Microsoftov Azure Custom Translator, ki je uporabljen v naši diplomski nalogi. Ker na sam model nimamo vpliva, se to delo v večini osredotoča na pridobivanje in pripravo podatkov. Z uporabo modelov LASER in Vecalign se iz profesionalno prevedenih besedil izloči in poravna vzporedne stavke. Z njimi se nauči dve različici prevajalnika po meri, ki temeljita na splošnem in tehnološkem osnovnem modelu. Z ocenami BLEU, chrF++ in BERTScore se naša modela primerja z drugimi možnostmi znotraj Azure ter eno vodilnih zunanjih storitev. Tehnološki model doseže rezultate, ki so primerljivi z zunanjim. Z najboljšim modelom izdelamo tudi preprosto prevajalniško aplikacijo.

Keywords

prevajanje;izdelava strojnega prevajalnika;označevalna industrija;Azure;oblačne storitve;univerzitetni študij;diplomske naloge;

Data

Language: Slovenian
Year of publishing:
Typology: 2.11 - Undergraduate Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [A. Zrimšek]
UDC: 004:81'322.4(043.2)
COBISS: 166032899 Link will open in a new window
Views: 42
Downloads: 9
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Developing a machine translation system for use in the labelling industry
Secondary abstract: Automatic machine translation can be an incredibly useful tool for companies whose employees are not fluent in English, because it can enable them to use software in their preferred language. While several companies offer machine translation services, these often do not reach the desired accuracy level when translating industry-specific texts. As a solution to this issue, some firms offer services that allow for translator tuning, using one’s own data. One of these is Microsoft’s Azure Custom Translator, which is the basis for this research paper. Since we cannot affect the model itself, this paper primarily focuses on gathering and processing the required data. Using LASER and Vecalign models, parallel sentences are extracted from professionally-translated texts and properly aligned. These are then used to train two separate versions of a custom translator, one based on a general baseline model, and the other on a technology baseline model. To evaluate our models, we employ the BLUE, chrF++, and BERTScore scoring systems to compare them with Azure’s other options, as well as one of the leading outside services. Upon completing our analysis, we conclude that our technology baseline model is comparable to the outside service. Finally, we use the best model to develop a simple translation app.
Secondary keywords: translation;translator;developing machine translation system;labeling industry;Azure;cloud services;computer science;diploma;Strojno učenje;Prevajanje in tolmačenje;Strojno prevajanje;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Bachelor thesis/paper
Study programme: 1000468
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 48 str.
ID: 21439470