Published: 2019-09-30
ISBN: 978-91-7929-997-2
ISSN: 1650-3686 (print), 1650-3740 (online)
As part of FNP Workshop Series, “Title Detection” is one of the two shared tasks proposed on Financial Document Structure Extraction. The objective of the task was to classify a given text block, that had been extracted from financial prospectuses in pdf format, as a title. Our DNN-based approach scored a weighted F1 of 97.16% on the test data.