id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-199267-cm6tqbzk	Wang, Zijie	Survive the Schema Changes: Integration of Unmanaged Data Using Deep Learning	2020-10-15		.txt	text/plain	8819	470	56	In this work, we propose to use deep learning to automatically deal with schema changes through a super cell representation and automatic injection of perturbations to the training data to make the model robust to schema changes. The contributions of this work include: (1) As to our best knowledge, we are the first to systematically investigate the application of deep learning and adversarial training techniques to automatically handle schema changes occurring in the data sources. A deep learning model, once trained, can handle most schema evolution without any human intervention, and does not require any data migration, or version management overhead. Our work has a potential to integrate data discovery and schema matching into a deep learning model inference process.	./cache/cord-199267-cm6tqbzk.txt	./txt/cord-199267-cm6tqbzk.txt