Turkish-Bert-NLP-Pipeline

Pipeline in a nutshell

In this project we aim at building Bert-base NLP pipeline for Turkish; Named Entity Recognition (Ner), Sentiment Analysis, Question Answering, Summarization and Text Categorization.

• For the pipeline in a nutshell, please check the file Turkish NLP Pipeline- Quick Start .ipynb
• For detailed application please check Turkish NLP Pipeline (For Detailed Presentation).ipynb

Details of Pipeline

1. Sentiment Analysis

• Pre-built model

• • https://huggingface.co/savasy/bert-base-turkish-sentiment-cased

• How to use it

• • https://github.com/savasy/TurkishSentimentAnalysis

2. NER Model

• Pre-built model

• • https://huggingface.co/savasy/bert-base-turkish-ner-cased

• How to use it

• • https://github.com/savasy/Turkish-Bert-Based-NERModel

3. Question Answering

• Pre-Built model

• • https://huggingface.co/savasy/bert-base-turkish-squad

• How to use it

• • https://huggingface.co/savasy/bert-base-turkish-squad

4. Text Summarization will be soon

5. Text Categorization

• Pre-built model

• • https://huggingface.co/savasy/bert-turkish-text-classification

• How to use it

• • https://github.com/savasy/TurkishTextClassification

These models are fined tuned based on Turkish-Bert model https://github.com/stefan-it/turkish-bert

All my models are found under

http://huggingface.co/savasy/

Plesa see Python Notebook file "Turkish NLP Pipeline with BERT.ipynb"