Wikimedia says the dataset hosted by Kaggle has been “designed with machine studying workflows in thoughts,” making it simpler for AI builders to entry machine-readable article knowledge for modeling, fine-tuning, benchmarking, alignment, and evaluation. The content material throughout the dataset is overtly licensed, and as of April fifteenth, consists of analysis summaries, quick descriptions, picture hyperlinks, infobox knowledge, and article sections — minus references or non-written components like audio recordsdata.
“Because the place the machine studying group comes for instruments and exams, Kaggle is extraordinarily excited to be the host for the Wikimedia Basis’s knowledge,” stated Kaggle partnerships lead Brenda Flynn. “Kaggle is happy to play a task in conserving this knowledge accessible, obtainable, and helpful.”
