Journals / Bilişim Teknolojileri Dergisi / 2019 / Cilt: 12 - Sayı: 3

Turkish News Articles Categorization Using Convolutional Neural Networks and Word2Vec

Türkçe Haber Metinlerinin Konvolüsyonel Sinir Ağları ve Word2Vec Kullanılarak Sınıflandırılması

Pages
219–228
Publication Date
2019-07-31
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

Due to the limited number of natural language processing operations in Turkish and the limited feature extraction methods in this area, the accuracy of the CNN models has increased by allowing the semantic values of the words to be included in the classification with the pre-trained Word2Vec network.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:In this study, a text classification study on the Turkish Text Classification 3600 (TTC-3600) dataset was conducted using Convolutional Neural Networks (CNN) and Word2Vec method and compared with the previous study using the same dataset.
📊
Key Findings & Data:In the study, two different CNN s were trained and tested on the TTC-3600 raw and stuck with Zemberek software.
💡
Conclusion & Impact:Due to the limited number of natural language processing operations in Turkish and the limited feature extraction methods in this area, the accuracy of the CNN models has increased by allowing the semantic values of the words to be included in the classification with the pre-trained Word2Vec network.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Turkish text categorization · convolutional neural networks · deep learning

Abstract

In this study, a text classification study on the Turkish Text Classification 3600 (TTC-3600) dataset was conducted using Convolutional Neural Networks (CNN) and Word2Vec method and compared with the previous study using the same dataset. In the study, two different CNN s were trained and tested on the TTC-3600 raw and stuck with Zemberek software. CNN and Word2Vec method showed better performance (93.3% accuracy) than classical statistical and machine learning based classification algorithms. Due to the limited number of natural language processing operations in Turkish and the limited feature extraction methods in this area, the accuracy of the CNN models has increased by allowing the semantic values of the words to be included in the classification with the pre-trained Word2Vec network.

Özet

Bu çalışmada, Konvolüsyonel Sinir Ağları (KSA) ve Word2Vec metodu kullanılarak Turkish Text Classification 3600 (TTC-3600) veri kümesi üzerinde metin sınıflandırma çalışması yapılmış ve aynı veri kümesi kullanılarak yapılan önceki çalışma ile kıyaslanmıştır. Çalışmada TTC-3600’ün ham ve Zemberek yazılımıyla gövdelenmiş halleri üzerinde iki farklı KSA eğitilmiş ve test edilmiştir. KSA ve Word2Vec metodu, klasik istatistiksel ve makine öğrenmesine dayalı sınıflandırma algoritmalarından daha iyi bir performans (%93,3 doğruluk) göstermiştir. Türkçe doğal dil işleme çalışmalarının azlığı ve bu alandaki özellik çıkarma yöntemlerinin limitli olması sebebiyle, kelimelerin semantik değerlerinin önceden eğitilmiş Word2Vec ağı ile sınıflandırmaya katılabilmesi KSA modellerinin doğruluk değerlerini arttırmıştır.

🔖

Cite This Article

ACI, Ç., ÇIRAK, A. (2019). Turkish News Articles Categorization Using Convolutional Neural Networks and Word2Vec. Bilişim Teknolojileri Dergisi, Cilt: 12 - Sayı: 3, 219–228. https://ajindex.com/articles/251737

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (16)

  1. Internet: World Internet Statistics. https://www.internetworldstats.com/stats.htm, 23.05.2019
  2. Internet: International Data Corporation. https://www.idc.com/, 23.05.2019
  3. N. Indurkhya, F.J. Damerau, Handbook of Natural Language Processing, Chapman & Hall/CRC, 2010
  4. C.D. Manning, P. Raghavan, H. Schütze, Introduction to information retrieval, Cambridge University Press, 2008
  5. E. Alpaydin, Machine learning: The New AI, The MIT Press, 2016
  6. H. Uğuz, “A two-stage feature selection method for text categorization by using information gain, principal component analysis and genetic algorithm”, Knowledge-Based Systems, 24(7), 1024–1032, 2011
  7. S. Jiang, G. Pang, M. Wu, L. Kuang, “An improved K-nearestneighbor algorithm for text categorization”, Expert Systems with Applications, 39(1), 1503–1509, 2012
  8. T. Jo, “Normalized table-matching algorithm as approach to text categorization”, Soft Computing, 19(4), 839–849, 2015
  9. B. Tang, H. He, P.M. Baggenstoss, S. Kay, “A Bayesian Classification Approach Using Class-Specific Features for Text Categorization”, IEEE Transactions on Knowledge and Data Engineering, 28(6), 1602–1606, 2016
  10. M.F. Amasyali, T. Yıldırım, “Automatic text categorization of news articles”, Proceedings of the IEEE 12th Signal Processing and Communications Applications Conference, Kusadasi, Turkey, 224–226, 28-30 April 2004
  11. R. Johnson, T. Zhang, “Effective Use of Word Order for Text Categorization with Convolutional Neural Networks”, arXiv:1412.1058v2, 2014
  12. X. Zhang, J. Zhao, Y. LeCun, “Character-level Convolutional Networks for Text Classification”, Advances in Neural Information Processing Systems, 649-657, Curran Associates Inc., 2015
  13. G. Biricik, “Sınıf Bilgisini Kullanan Boyut İndirgeme Yöntemlerinin Metin Sınıflandırmadaki Etkilerinin Karşılaştırılması”, 20. IEEE Sinyal İşleme ve İletişim Uygulamaları Kurultayı, Muğla, Türkiye, 1–4, 2012
  14. A. Haltaş, A. Alkan, M. Karabulut, “Metin Sınıflandırmada Sezgisel Arama Algoritmalarının Performans Analizi”, Journal of the Faculty of Engineering and Architecture of Gazi University, 30(3), 2015
  15. D. Kılınç, E. Borandağ, F. Yücalar, V. Tunalı, M. Şimşek, A. Özçift, “KNN Algoritması ve R Dili ile Metin Madenciliği Kullanılarak Bilimsel Makale Tasnifi”, Marmara Fen Bilimleri Dergisi, 28(3), 89–94, 2016
  16. Ç. Çatal, K. Erbakırcı, Y. Erenler, “Computer-based Authorship Attribution for Turkish Documents”, Turkish Symposium on Artificial Intelligenc