Language is the primary medium used by human beings to convey their thoughts, ideas, feelings, and information. There are many languages in the world each with its own unique complexities. Therefore, language barrier among people is rapidly increased, and the language complexity has been become an unsolved problem in linguistics. However, this language complexity can be reduced using the technology named “Machine Translation” which is one of the areas in Natural Language Processing. It is a computer-aided machine that can translate one language into another language without any intervention of humans. Although, there are many machine translation systems in the world to translate different language pairs, this process still remained as a complex process due to various reasons. The main problem behind this situation is there is no any universal language interlingua model for machine translation to represent and model language information of a particular language that could use for any translation. As the solution, it is recommended to design and develop a universal language model that could facilitate machine translation. As the first step of this research, developing a universal morphological model for English language is proposed that can be used to generate appropriate target morphological model for any language. The main aim of this article is to study and compare the morphological differences and complexities of two non-related language pairs namely English and Sinhala to design and develop this universal morphological model. Therefore, these two selected languages were deeply studied and analyzed in morphological point of view and many promising differences have been identified in respect to grammar structures, parts-of-speech, inflectional categories and etc.
@artical{m13112024ijcatr13111010,
Title = "Study Morphological Complexity of Non-Related Languages to Build a Universal Morphological Model for Machine Translation ",
Journal ="International Journal of Computer Applications Technology and Research (IJCATR)",
Volume = "13",
Issue ="11",
Pages ="65 - 72",
Year = "2024",
Authors ="MAST Goonatilleke, B Hettige, AMRR Bandara"}