/Arabic-Dialects-Identification

The Dataset and the dialect identification problem were addressed by Qatar Computing Research Institute. I implemented a data-scraper, data-preprocessor (Regex and NLTK), data-modeling (SVM with TF-IDF and MarBert using HuggingFace) and deployment using FlaskAPIs locally.

Primary LanguageJupyter NotebookThe UnlicenseUnlicense

Watchers