The Sindhi News Corpus (Daily Wakka) is a Sindhi-language newspaper dataset collected from Daily Wakka for Natural Language Processing (NLP) research. This dataset is designed to support research in low-resource language processing, particularly for Sindhi.

Each entry contains:

Category — News section
Title — News headline
Content — Full news article text

The dataset is structured in Parquet (.parquet) format (UTF-8 encoding).

Use Cases

Text Classification
News Summarization
Headline Generation
Language Modeling
Named Entity Recognition (NER)
Question Answering
Information Retrieval
Low-Resource NLP Research

Format

Parquet file with three columns:

Category
Title
Content

Notes

Encoding: UTF-8
Cleaned and structured for NLP tasks
Efficient columnar storage using Parquet
Suitable for training and fine-tuning transformer-based models

Author

Danish Mahdi
Hugging Face: https://huggingface.co/DanishMahdi

Downloads last month: 6

Total file size:

795 kB