Hội thảo khoa học quốc gia "A SENSE TAGGING ALGORITHM USING UNSUPERVISED METHO"

Tagging Sense on multiple languages has been studying with many popular languages as English, German, Japanese, French, etc. However, semantic labelling task for unpopular languages as Vietnamese are still many limitations, especially for performing the sense similarities on bilingual English-Vietnamese. In this article, we propound a solution for tagging semantic labels automatically in the English-Vietnamese bilingual corpus to take full benefits of the translations of cross-language lexicon, but it also conserves the kernel constituents of its sense. This architecture has used corpus from the web to build sets combined with the possibility to combine different meaning of words found in the corpus, and it has also used an unsupervised algorithm to tag the sense label in English, which depended upon sense similarities cross English - Vietnamese corpus. Then, this model will automatically project labels from English to Vietnamese via available links that have been recorded previously

 Chi tiết bài nghiên cứu: https://vap.ac.vn/Portals/0/TuyenTap/2020/4/18/fa7f0835392b41f4a4751e1593be09f5/2-92.pdf

 Cùng chia sẻ để đưa công nghệ ngôn ngữ Việt lên tầm cao mới!