Swahili News Dataset

Swahili, or Kiswahili, is a Bantu language spoken by millions of people, primarily in East Africa. It serves as the official language of several countries, including Kenya, Tanzania, Uganda, and the Democratic Republic of Congo, and is widely used as a lingua franca in the region. Swahili is known for its rich linguistic history, influenced by Arabic, Persian, Portuguese, and other African languages due to centuries of trade and cultural exchange along the East African coast. It is also the language of Swahili news and media outlets. It is a vital medium for delivering up-to-date information on various topics to Swahili-speaking populations, thus playing a central role in the region's communication and information dissemination. This dataset combines the dataset published by David on Zenodo (https://zenodo.org/record/4300294), and the dataset scraped from different online sources. It contains 30K+ swahili news texts (entry sentences), each labelled with one of six news categories: kitaifa ("national"), michezo ("sports"), burudani ("entertainment"), uchumi ("economy"), kimataifa ("international"), and afya ("health"). The dataset can be used for Swahili news classification. We collected more data because the original data set from David's data was imbalanced, and a more extensive dataset was needed to get better classification models.