http://www1.lainac.c.u-tokyo.ac.jp/research/seminars/bdnlp
N
estor Alvaro (Data Scientist at Deloitte Digital, Spain)
Developing an annotation framework to extract insights from news

Date: June 7, 2019 (Friday)
Time: 15:00 - 17:00 (Japan Time)
Venue: UTokyo, Komaba Campus, Building 2, Room 303



In order to process the information contained in the texts we should focus on obtaining certain types of information moving from identifying keywords, to discovering specific insights. For this, we should focus on what to extract, for which we combine human annotators, who can identifying such insights in the text, and then move to an automatic process where machine learning algorithms identify such insights by learning the patterns in the annotations provided by the human annotators. In this workshop we will see how to prepare the annotation guidelines for human annotators, so they can annotate these insights, allowing us to use such insights at a later stage to discover insights and train machine learning algorithms.