Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Context Aware Lemmatization and Morphological Tagging Method in Turkish

Cagri Sayallar

TL;DR

The paper tackles Turkish lemmatization and morphological tagging with explicit word-meaning awareness to address Turkish’s agglutinative morphology. It introduces two context-sensitive neural architectures that fuse spelling information from Bi-LSTM with word meaning from Turkish DistilBERT, and explores two model variants: a separate model and a sequenced model that uses predicted morphological tags as input. Trained and evaluated on Universal Dependencies Turkish IMST and PUD datasets against the SIGMORPHON 2019 baselines, the proposed methods deliver superior results in most metrics, with the sequenced model often performing best. This work fills a gap by delivering the first Turkish-focused, meaning-aware lemmatization approach and demonstrates the practical impact of incorporating word meaning into morphology-aware NLP for Turkish.

Abstract

The smallest part of a word that defines the word is called a word root. Word roots are used to increase success in many applications since they simplify the word. In this study, the lemmatization model, which is a word root finding method, and the morphological tagging model, which predicts the grammatical knowledge of the word, are presented. The presented model was developed for Turkish, and both models make predictions by taking the meaning of the word into account. In the literature, there is no lemmatization study that is sensitive to word meaning in Turkish. For this reason, the present study shares the model and the results obtained from the model on Turkish lemmatization for the first time in the literature. In the present study, in the lemmatization and morphological tagging models, bidirectional LSTM is used for the spelling of words, and the Turkish BERT model is used for the meaning of words. The models are trained using the IMST and PUD datasets from Universal Dependencies. The results from the training of the models were compared with the results from the SIGMORPHON 2019 competition. The results of the comparisons revealed that our models were superior.

Context Aware Lemmatization and Morphological Tagging Method in Turkish

TL;DR

The paper tackles Turkish lemmatization and morphological tagging with explicit word-meaning awareness to address Turkish’s agglutinative morphology. It introduces two context-sensitive neural architectures that fuse spelling information from Bi-LSTM with word meaning from Turkish DistilBERT, and explores two model variants: a separate model and a sequenced model that uses predicted morphological tags as input. Trained and evaluated on Universal Dependencies Turkish IMST and PUD datasets against the SIGMORPHON 2019 baselines, the proposed methods deliver superior results in most metrics, with the sequenced model often performing best. This work fills a gap by delivering the first Turkish-focused, meaning-aware lemmatization approach and demonstrates the practical impact of incorporating word meaning into morphology-aware NLP for Turkish.

Abstract

The smallest part of a word that defines the word is called a word root. Word roots are used to increase success in many applications since they simplify the word. In this study, the lemmatization model, which is a word root finding method, and the morphological tagging model, which predicts the grammatical knowledge of the word, are presented. The presented model was developed for Turkish, and both models make predictions by taking the meaning of the word into account. In the literature, there is no lemmatization study that is sensitive to word meaning in Turkish. For this reason, the present study shares the model and the results obtained from the model on Turkish lemmatization for the first time in the literature. In the present study, in the lemmatization and morphological tagging models, bidirectional LSTM is used for the spelling of words, and the Turkish BERT model is used for the meaning of words. The models are trained using the IMST and PUD datasets from Universal Dependencies. The results from the training of the models were compared with the results from the SIGMORPHON 2019 competition. The results of the comparisons revealed that our models were superior.
Paper Structure (10 sections, 3 figures, 2 tables)

This paper contains 10 sections, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Method
  4. Encoder
  5. Decoder
  6. Experiments and Results
  7. Dataset
  8. Training Parameters
  9. Tests
  10. Conclusion

Figures (3)

  • Figure 1: Separate Model and Sequenced Model.
  • Figure 2: Encoder scheme of the model used in the lemmatization and morphological tagging model. The arrow represents the layer output. The dashed arrow represents the hidden state output of the last Bi-LSTM layer.
  • Figure 3: Decoder scheme of the model used in the lemmatization and morphological tagging model. The arrow represents the layer output. The dashed arrow represents the hidden state output of the last Bi-LSTM layer.