تم استخدام تحلل الطابع الصيني كميزة لتعزيز نماذج الترجمة الآلية (MT)، والجمع بين المتطرفين في طرازات حرف مستوى الكلمة.حققت العمل الحديث في الأيديوجراف أو تضمين مستوى السكتة الدماغية.ومع ذلك، تبقى الأسئلة حول مستويات التحلل المختلفة من تمثيلات الأحرف الصينية، والراديكالية والسكتات الدماغية، والأمن الأكون مناسبة لجبل.للتحقيق في تأثير تضمين التحلل الصيني بالتفصيل، أي المستويات الجذعية والسكتة الدماغية والسكتة الدماغية، ومدى جودة تحلل هذه التحلل معنى تسلسل الأحرف الأصلية، نقوم بإجراء تحليل مع كل من التقييم الآلي والإنساني ل MT.علاوة على ذلك، يمكننا التحقيق في ما إذا كان يمكن أن يعزز مزيج التعبيرات المتعددة الكلمة المتحللة (MWES) التعلم النموذجي.شهدت تكامل MWE في MT أكثر من عقد من الاستكشاف.ومع ذلك، لم يتم استكشاف mwes المتحللة سابقا.
Chinese character decomposition has been used as a feature to enhance Machine Translation (MT) models, combining radicals into character and word level models. Recent work has investigated ideograph or stroke level embedding. However, questions remain about different decomposition levels of Chinese character representations, radical and strokes, best suited for MT. To investigate the impact of Chinese decomposition embedding in detail, i.e., radical, stroke, and intermediate levels, and how well these decompositions represent the meaning of the original character sequences, we carry out analysis with both automated and human evaluation of MT. Furthermore, we investigate if the combination of decomposed Multiword Expressions (MWEs) can enhance the model learning. MWE integration into MT has seen more than a decade of exploration. However, decomposed MWEs has not previously been explored.
References used
https://aclanthology.org/
This paper presents an attempt at multiword expressions (MWEs) discovery in the Persian language. It focuses on extracting MWEs containing lemmas of a particular group: loanwords in Persian and their equivalents proposed by the Academy of Persian Lan
Sememes are defined as the atomic units to describe the semantic meaning of concepts. Due to the difficulty of manually annotating sememes and the inconsistency of annotations between experts, the lexical sememe prediction task has been proposed. How
Character-based word-segmentation models have been extensively applied to agglutinative languages, including Thai, due to their high performance. These models estimate word boundaries from a character sequence. However, a character unit in sequences
In this paper, we present the systems submitted by our team from the Institute of ICT (HEIG-VD / HES-SO) to the Unsupervised MT and Very Low Resource Supervised MT task. We first study the improvements brought to a baseline system by techniques such
Recent state-of-the-art (SOTA) effective neural network methods and fine-tuning methods based on pre-trained models (PTM) have been used in Chinese word segmentation (CWS), and they achieve great results. However, previous works focus on training the