KELİME TEMSİLLERİ İÇİN TEST PERFORMANSINI GELİŞTİRMEYE YÖNELİK EŞDİZİMLİLİK AĞIRLIKLARININ SEÇİMİ

Bu çalışma, matematiksel kelime temsillerinin belirli bir görev için performanslarının en iyilenmesi problemini yeniden ele almaktadır. Sayma tabanlı (kelimelerin eşdizimlilik istatistiklerini hesaba katan) kelime temsili oluşturma yöntemlerinde klasik olarak kullanılan sayma ağırlıkları yerine yenilikçi ağırlıklar önererek analoji ve benzerlik bulma görevlerinde performans artışı sağlamak hedeflenmektedir. Çalışma dili olarak Türkçe seçilmiş, derlem oluşturulurken Türkçe’ye has ek-kök yapıları ek alan her kelime yeni bir kelime gibi kabul edilecek şekilde yorumlanmıştır. Önerilen eşdizimlilik ağırlıklarının performansı değişen parametreye göre analiz edilerek sonuçlar çalışma içerisinde paylaşılmıştır.

Anahtar Kelimeler:

Kelime temsilleri, Doğal dil işleme, İstatistiksel dilbilimi

Co-occurrence Weight Selection for Word Embeddings to Enhance Test Performance

This study revisits the problem of maximizing the performance of mathematical word representations for a given task. It is aimed to improve performance in analogy and similarity tasks by suggesting innovative weights instead of the counting weights used conventionally in counting-based methods of generating word representations (adding the statistics of word co-occurrences to the account). The language of study was selected as Turkish. The root structures of Turkish words were managed during the compilation of corpus such that each word having a suffix was considered as a new word. The performance of the proposed co-occurrence weights are analyzed with respect to the varying parameter and the results are presented within the paper.

Keywords:

Word embeddings, Natural language processing, Statistical linguistics,

PDF

___

Bahdanau, D., Cho, K. and Bengio, Y. (2014). Neural machine translation by jointly learning to align and translate. arXiv preprint arXiv: 1409.0473.
Bengio, Y., Ducharme, R., Vincent, P., and Jauvin C. (2003). A neural probabilistic language model. Journal of machine learning research, 1137 – 1155. doi: 10.1162/153244303322533223
Bojanowski, P., Grave, E., Joulin, A., ve Mikolov, T. (2016). Enriching word vectors with subword information. arXiv preprint arXiv:1607.04606.
Faruqui, M., Dodge, J. , Jauhar, S. K., Dyer, C., Hovy, E. ve Smith, N. A. (2014) Retrofitting word vectors to semantic lexicons, arXiv preprint arXiv:1411.4166. doi: 10.3115/v1/N15-1184
Firth, J. R., (1957). A synopsis of linguistic theory 1930-1955. In Studies in linguistic analysis, 1-32. Oxford:Blackwell.
Huth, A.G., de Heer, W.A., Griffiths, T.L., Theunissen, F.E. and Gallant, J.L. (2016) Natural speech reveals the semantic map that tile human cerebral cortex. Nature, vol. 532, no. 7600, 453 – 458. doi:10.1038/nature17637
Karpathy, A. and Fei-Fei, L. (2016). Deep visual-semantic alignments for generating image descriptions. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39 (4), 664-676. doi: 10.1109/TPAMI.2016.2598339
Krizhevsky, A., Sutskever, I., and Hinton, G. E. (2012). Imagenet classification with deep convolutional neural netwroks. In Advances in neural information processing systems, 1097-1105. doi: 10.1145/3065386
Le, Q. V. ve Mikolov, T. (2014) Distributed representations of sentences and documents, ICML, vol. 14, 1188–1196.
Luong, T., Socher, R. ve Manning, C.D. (2013) Better word representations with recursive neural networks for morphology, CoNLL, 104–113.
Mikolov, T., Karafiat, M., Burget, L., Cernocky, J. and Khudanpur S. (2010) Recurrent neural network based language model, Interspeech, Vol 2, 3.
Mikolov, T., Chen, K., Corrado, G. ve Dean, J. (2013a) Efficient estimation of word representations in vector space, arXiv preprint arXiv:1301.3781.
Mikolov, T., Le, Q.V. and Sutskever, I. (2013b) Exploiting similarities among languages for machine translation. arXiv preprint arXiv:1309.4168.
Mikolov, T., Sutskever, I, Chen, K., Corrado, G.S. ve Dean J. (2013c) Distributed representations of words and phrases and their compositionality, Advances in neural information processing systems, 3111–3119.
Mnih, A., ve Hinton, G., (2007) Three new graphical models for statistical language modelling, Proceedings of the 24th International Conference on Machine Learning. ACM, 641–648. doi:10.1145/1273496.1273577
Pennington, J., Socher, R. ve Manning, C.D. (2014) Glove: Global vectors for word representation, Empirical Methods in Natural Language Processing (EMNLP), 1532–1543. doi: 10.3115/v1/D14-1162
Ravichandran, D., Pantel, P. ve Hovy, E. (2005) Randomized algorithms and nlp: Using locality sensitive hash function for high speed noun clustering, Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics, Association for Computational Linguistics, 622–629. doi: 10.3115/1219840.1219917
Salton, G., Wong, A., and Yang, C.S. (1975) A vector space model for automatic indexing. Communications of the ACM, 18(11), 613-620. doi: 10.1145/361219.361220
Socher, R., Perelygin, A., Wu, J.Y., Chuang, J., Manning, C.D., Ng, A.Y., ve Potts C., (2013) Recursive deep models for semantic compositionality over a sentiment treebank, Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 1631-1642.
Şenel, L.K., Yücesoy, V., Koç, A. and Çukur, T. (2017) Measuring cross-lingual semantic similarity across European languages. In International Conference on Telecommunications and Signal Processing (TSP), 359-363. doi: 10.1109/TSP.2017.8076005
Şenel, L. K., Yücesoy, V., Koç, A., Çukur, T. (2017b). Semantic similarity between Turkish and european languages using word embeddings. 25th Signal Processing and Communications Applications Conference, 1-4. doi: 10.1109/SIU.2017.7960365
Tang, D., Wei, F., Yang, N., Zhou, M., Liu, T. and Qin, B. (2014) Learning sentiment specific word embedding for twitter sentiment classification. ACL, 1555 – 1565. doi: 10.3115/v1/P14-1146
Yücesoy, V., Koç A. (2017). Effect of co-occurrence weighting to English word embeddings. 25th Signal Processing and Communications Applications Conference, 1-4. doi: 10.1109/SIU.2017.7960385