Automatic keyword annotation using newspapers

An index term that affects the precision of retrieval is important for a retrieval system. Retrieval systems usually use a method that automatically extracts a search word from a document, but a method that gives keywords as search words manually is also necessary for newspapers. Therefore, we propose the Nikkei annotator system, which is based on the model of the human brain and learns patterns of past keyword annotation and automatically outputs keywords that are preferred by us, and verify the superiority of this system by comparing it with comparable approaches.