Gizli Markov Modeli Kullanılarak Özbek Dilinde Sözcüklerin Nitelikleri ile Etiketlemesi

Öz

Markov modelleri, doğal dil işlemede en yaygın kullanılan makine öğrenmesi yöntemlerinden biridir. Markov zinciri ve gizli Markov modeli, dinamik sistemleri modellemek için kullanılan stokastik (rastlantısal) yöntemdir ve sistemin mevcut durumunu, önceki durumlara dayanarak tahmin eder. Tümcelerin oluşturulmasında doğru bir sözcük dizisi oluşturan Markov zinciri, DDİ çalışmalarında yaygın olarak kullanılmaktadır. Ayrıca bir tümcedeki Varlık Adlarını (NER : VAT) tanımlamak ve gizli Markov modeline dayalı Sözcükleri Nitelikleri ile Etiketlemek (SNE) için kullanılır. Markov modeline dayanarak, gizli etiketler derlemdeki etiketli sözcüklere göre tahmin edilir. Bu makale, Özbek dilinin etiketli derlemini temel alan bir Gizli Markov modeli kullanarak belirli sözcüklerin otomatik SNE etiketlemesine yönelik yöntemler ve algoritmaları ve bunların uygulanmasını sunmaktadır.

Anahtar Kelimeler

• SNE (POS) etiketleme
• Markov modeli
• Gizli Markov modeli
• Markov zinciri
• Rastantısal Yöntemler
• DDİ (NLP)
• Sözcük Sınıfları
• Eşsesli Çözümleme
• Geçiş Olasılığı
• Yayılma Olasılığı
• Viterbi Algoritması.

Tagging Words with Their Attributes in Uzbek Language Using the Hidden Markov Model

Öz

Markov models are one of the most widely used machine learning methods in natural language processing. Markov chain and hidden Markov model are stochastic (random) methods used to model dynamic systems and predict the current state of the system based on previous states. Markov chain, which creates a correct sequence of words in the creation of sentences, is widely used in NLP studies. It is also used to identify NERs in a phrase and for POS tagging based on hidden Markov model. Based on the Markov model, latent labels are predicted based on the labeled words in the language corpus. This article presents methods and algorithms for automatic POS tagging of a given phrase using a hidden Markov model based on a labeled corpus of the Uzbek language

Anahtar Kelimeler

• POS tagging
• hidden Markov model
• Markov chain
• Hidden Markov Models
• Stochastic methods
• NLP
• Word classes
• Homophone analysis
• Transition probability
• Propagation probability
• Viterbi algorithm

Kaynakça

1. Rohman, Y. A., Kusumaningrum, R., Twitter Storytelling Generator Using Latent Dirichlet Allocation And Hidden Markov Model POS-TAG (Part-Of-Speech Tagging). ICICOS 2019 - 3rd International Conference on Informatics and Computational Sciences: Accelera-ting Informatics and Computational Research for Smarter Society in The Era of Industry 4.0, Proceedings. https://doi.org/10.1109/ICICoS48119.2019.8982411, (2019).
2. Muljono, U., Supriyanto, C., Morphology Analysis For Hidden Markov Model Based Indonesian Part-of-Speech Tagger. Proceedings - 2017 1st International Conference on Informatics and Computational Sciences, ICICoS 2017, 2018-January. https://doi.org/10.1109/ICICOS.2017.8276368, (2017).
3. Xusainova, Z. Y. , NLP: Tokenizatsiya, Stemming, Lemmatizatsiya Va Nutq Qismlarini Teglash. O‘Zbek Amaliy Filologiyasi İstiqbollari / Respublika İlmiy-Amaliy Konferensiya To‘Plami. Elektron nashr / ebook. – Toshkent: ToshDOʻTAU, 26.10.2022. 159-163 b.
4. Baishya, D., Baruah, R., Resource Languages with Improved Hidden Markov Model and Deep Learning. International Journal of Advan-ced Computer Science and Applications, 12(10). https://doi.org/10.14569/IJACSA.2021.0121011, (2021).
5. Pattnaik, S., Nayak, A. K., Patnaik, S., A Semi-Supervised Learning Of HMM To Build A POS Tagger for a Low Resourced Language. Journal of Information and Communication Convergence Engineering, 18(4). https://doi.org/10.6109/jicce.2020.18.4.207, (2020).
6. Elov, B., Hamroyeva, Sh., Axmedova, X., Methods for Creating a Morphological Analyzer, 14th International Conference on Intelligent Human Computer Interaction, IHCI 2022, 19-23 October 2022, Tashkent. (2022).
7. Assunção, J., Fernandes, P. Lopes, L., Language Independent Pos-Tagging Using Automatically Generated Markov Chains. Proceedings of the International Conference on Software Engineering and Knowledge Engineering, SEKE, 2019-July. https://doi.org/10.18293/SEKE2019-097, (2019).
8. Talarico, E., Leão, W., Grana, D., Comparison of Recursive Neural Network and Markov Chain Models İn Facies Inversion. Mathematical Geosciences, 53(3). https://doi.org/10.1007/s11004-020-09914-w, (2021).

9. Myers, D. S., Wallin, L., Wikström, P., An Introduction to Markov Chains and Their Applications Within Finance. Mathematical Sciences-Chalmers University of Technology and University, (2017).
10. Cahyani, D. E., Mustikaningtyas, W., Indonesian Part of Speech Tagging Using Maximum Entropy Markov Model on Indonesian Manually Tagged Corpus. IAES International Journal of Artificial Intelligence, 11(1). https://doi.org/10.11591/ijai.v11.i1.pp336-344 , (2022).
11. Jurafsky, D., Martin, J., Speech and Language Processing. In Speech and Language Processing. (Vol. 3). (2014).
12. Cing, D. L., Soe, K. M. , Improving Accuracy of Parto-of-Speech (POS) Tagging Using Hidden Markov Model and Morphological Analysis For Myanmar Language. International Journal of Electrical and Computer Engineering, 10(2). https://doi.org/10.11591/ijece.v10i2.pp2023-2030 , (2020).
13. Kadim, A. Lazrek, A., Parallel HMM-Based Approach for Arabic Part of Speech Tagging. International Arab Journal of Information Tech-nology, 15(2). (2018).
14. Suleiman, D., Awajan, A., Etaiwi, W., The Use of Hidden Markov Model in Natural ARABIC Language Processing: A Survey. Procedia Computer Science, 113. https://doi.org/10.1016/j.procs.2017.08.363 , (2017).
15. Kumar, A.. Katiyar, V., Kumar, P., A Comparative Analysis of Pre-Processing Time in Summary Of Hindi Language Using Stanza and Spacy. IOP Conference Series: Materials Science and Engineering, 1110(1). https://doi.org/10.1088/1757-899x/1110/1/012019, (2021).
16. Atmakuri, S., Shahi, B., Ashwath Rao, B., Muralikrishna, S. N. , A Comparison of Features For POS Tagging in Kannada. International Journal of Engineering and Technology(UAE), 7(4). https://doi.org/10.14419/ijet.v7i4.14900 (2018).
17. Chiche, A., Kadi, H., Bekele, T., A Hidden Markov Model-Based Part of Speech Tagger for Shekki’noono Language. International Journal of Computing, 2021(4). https://doi.org/10.47839/ijc.20.4.2448, (2021).
18. Al-Anzi, S., A.Zeina, D., Statistical Markovian Data Modeling for Natural Language Processing. International Journal of Data Mining & Knowledge Management Process, 7(1). https://doi.org/10.5121/ijdkp.2017.7103 , (2017).
19. Kumawat, D., Jain, V., POS Tagging Approaches: A Comparison. International Journal of Computer Applications, 118(6). https://doi.org/10.5120/20752-3148, (2015).
20. Jassim, A. K., Al-Bayaty, B. , A Stochastic Approach to Identify POS in Iraqi National Song Using N-Iterative HMM Using Agile Approach. IOP Conference Series: Materials Science and Engineering, 1094(1). https://doi.org/10.1088/1757-899x/1094/1/012019, (2021).
21. Hadni, M., Alaoui Ouatik, S., Lachkar, A., Meknassi, M., Hybrid Part-of-Speech Tagger For Non-Vocalized Arabic Text. International Journal on Natural Language Computing, 2(6). https://doi.org/10.5121/ijnlc.2013.2601, (2013).
22. Bohnet, B., McDonald, R., Simões, G., Andor, D., Pitler, E., May-nez, J., Morphosyntactic Tagging with a Meta-Bilstm Model Over Context Sensitive Token Encodings. ACL 2018 - 56th Annual Meeting of the Association for Computational Linguistics, Proceedings of the Conference (Long Papers), 1. https://doi.org/10.18653/v1/p18-1246, (2018).
23. Khoe, Y. H., Reproducing a Morphosyntactic Tagger With a Meta-Bilstm Model Over Context Sensitive Token Encodings. LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings. (2020).
24. Elov, B., Axmedova, X., Determining Homonymy Using Statistical Methods. Computational Models and Technologies, Uzbekistan-Malaysia international conference, September 16-17th, 2022, Tash-kent. (2022).
25. Elov, B., Axmedova, X., Business Process Modeling That Distinguis-hes Homonymy Within Three Parts Of Speechs in Uzbek Language. 7th International Conference on Computer Science and Engieering, 14-16 September 2022, Turkey, (2022).
26. Zucchini, W., MacDonald, I.L., Langrock, R. (2016). Hidden Markov Models for Time Series: An Introduction Using R, Second Edition (2nd ed.). Chapman and Hall/CRC. https://doi.org/10.1201/b20790.
27. Al-Anzi, F. ve AbuZeina, D., A Survey of Markov Chain Models in Linguistics Applications. 53-62. 10.5121/csit.2016.61305. (2016).
28. Kadim, A. ve Lazrek, A., Bidirectional HMM-based Arabic POS tag-ging. International Journal of Speech Technology. 19. 10.1007/s10772-015-9303-7. (2016).