O‘ZBEK TILIDAGI GAPLARNI PEREFRAZ QILISH BOSQICHLARI VA MUAMMOLARI
Keywords:
Perefraz, taksonomiya, NLP, sinonimlar, semantika, neyron modellar.Abstract
Ushbu tadqiqot perefrazlar taksonomiyasi tushunchasi va uning leksik, sintaktik hamda semantik darajalardagi ierarxik tuzilishi batafsil yoritilgan. Semantik perefrazlar esa mazmunni umumlashtirish, aniqlashtirish, tafsilot qo‘shish yoki yashirin qarama-qarshilik yaratish asosida tahlil qilingan. Shuningdek, perefraz yaratishning an’anaviy (sinonimlarni almashtirish, WordNet/tezaurusga asoslangan) va neyron modellar (Seq2Seq, LSTM) asosidagi usullari hamda ularning afzallik va cheklovlari ko‘rib chiqilgan. O‘zbek tilidagi so‘z, ibora va gap darajasidagi perefrazlash imkoniyatlari misollar orqali yoritilib, perefraz generatsiyasi NLP ilovalari - savol-javob tizimlari, mashina tarjimasi va axborot izlashda muhim ahamiyatga ega ekani asoslab berilgan.
References
1. J Edward Hu, Rachel Rudinger, Matt Post, and Benjamin Van Durme. Parabank: Monolingual bi text generation and sentential paraphrasing via lexically constrained neural machine translation. arXiv preprint, 2019.
2. Lin, D., Pantel, P.: Discovery of inference rules for question-answering. Natural Language Engineering 7(4), 343–360 (2001)
3. Zhou, J., & Bhat, S. (2021). Paraphrase Generation: a survey of the state of the art. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing 5075-5086-p. https://doi.org/10.18653/v1/2021.emnlp-main.414
4. Prakash, A., Hasan, S. A., Lee, K., Datla, V. V., Qadir, A., Liu, J., & Farri, O. (2016). Neural Paraphrase Generation with Stacked Residual LSTM Networks. International Conference on Computational Linguistics, 2923–2934. Retrieved from https://www.aclweb.org/anthology/C16-1275.pdf
5. Lin, Z., & Wan, X. (2021). Pushing Paraphrase Away from Original Sentence: A Multi-Round Paraphrase Generation Approach. Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, 1548-1557-p. https://doi.org/10.18653/v1/2021.findings-acl.135
