Document classification


A large video gaming company stores 6-million documents in their knowledge management platform. They need their  employees to easily access these documents, for them to capitalize on technical, gameplay & gamer experience best practices.  They need their documents to be better categorized.


The video gaming company needed support to build the dataset to be used to train a Natural Language Processing model that

automatically classifies documents within their Knowledge Management platform.


Ingedata built a team of Native-English annotators to review and tag documents. An agile production process was set to  streamline production and achieve quality. The annotation process was purpose-designed to meet Recall and Precision targets.