turkish-nlp-suite/Havadis

Dataset Card for Havadis Havadis is a high quality and large Turkish news corpus, indeed the largest Turkish news corpus ever. This corpus is scraped from online news sebsites and includes text from popular newspapers such as CNN Türk Habertürk Hürriyet Millyet NTV Posta Sabah Star Sözcü Takvim . The instances are first crawled from the corresponding websites, then went throught an extensive cleaning process. We eliminated instances that are too short, too repetetive… See the full description on the dataset page: https://huggingface.co/datasets/turkish-nlp-suite/Havadis.