Published: 3 April 2023| Version 1 | DOI: 10.17632/dntxt73dm6.1
Soran Badawi


Kurdish Emotional Dataset is a new dataset for Kurdish language which includes over 77000 texts. This dataset contains four labels: fear, joy, sad, and surprise. The dataset is intended to be used for training and evaluating models for emotion recognition in the Kurdish language. The dataset is composed of tweets of Kurdish speakers expressing different emotions. Each tweets is labeled according to the emotion expressed. The dataset has been carefully curated and reviewed to ensure high quality and accuracy. We believe that this dataset will be a valuable resource for researchers and developers working on emotion recognition applications for the Kurdish language. We encourage researchers and developers to use this dataset in their work, and we welcome feedback and contributions to further improve the dataset.



Machine Learning, Emotion, Kurd, Deep Learning, Sentiment Analysis