Stručni rad
https://doi.org/10.36978/cte.5.2.4
Analysis of the automatic statistical machine translation system model
Ivan Dunđer
; Filozofski fakultet Sveučilišta u Zagrebu
Sažetak
Automatic machine translation is an increasingly popular research topic in science and various scientific disciplines, such as information and communication sciences, computer science, computational linguistics etc. The reason for this is primarily that today it enables unavoidable communication and fast transfer of information between different natural languages. This is especially important for less spoken languages such as Croatian, for which there are still not enough software tools and digital resources that are needed for developing specialized and quality machine translation systems that would be optimized for use in one specific area. The evermore faster growth of data and the growing need of various stakeholders in the sectors of industry, economy, science, but also in peoples’ everyday life imply the motivation for the systematic and organized development and subsequent adaptation of automatic machine translation systems for different language pairs. Since machine translations are not perfect, it is important to apply methods for computationally generating translations of an acceptable level of quality that depends on the task itself and the scope of implementation of the machine translation system. In this paper, the model of a system for automatic statistical machine translation, its components and the role and significance of individual elements within the model are analyzed.
Ključne riječi
automatic machine translation; statistical machine translation model; language technologies; natural language processing; information and communication sciences
Hrčak ID:
271055
URI
Datum izdavanja:
22.12.2021.
Posjeta: 611 *