caa a22 4500 445379642 CHVBK 20180317143013.0 cr unu---uuuuu 170323e20110601xx s 000 0 eng 10.1007/s10032-011-0150-z doi (NATIONALLICENCE)springer-10.1007/s10032-011-0150-z Robust named entity detection from optical character recognition output [Elektronische Daten] [Krishna Subramanian, Rohit Prasad, Prem Natarajan] In this paper, we focus on information extraction from optical character recognition (OCR) output. Since the content from OCR inherently has many errors, we present robust algorithms for information extraction from OCR lattices instead of merely looking them up in the top-choice (1-best) OCR output. Specifically, we address the challenge of named entity detection in noisy OCR output and show that searching for named entities in the recognition lattice significantly improves detection accuracy over 1-best search. While lattice-based named entity (NE) detection improves NE recall from OCR output, there are two problems with this approach: (1) the number of false alarms can be prohibitive for certain applications and (2) lattice-based search is computationally more expensive than 1-best NE lookup. To mitigate the above challenges, we present techniques for reducing false alarms using confidence measures and for reducing the amount of computation involved in performing the NE search. Furthermore, to demonstrate that our techniques are applicable across multiple domains and languages, we experiment with optical character recognition systems for videotext in English and scanned handwritten text in Arabic. Springer-Verlag, 2011 Optical character recognition nationallicence Hidden Markov Model nationallicence Information extraction nationallicence Named entity detection nationallicence OCR : Optical character recognition nationallicence HMM : Hidden Markov model nationallicence ROC : Receiver operating characteristics nationallicence Subramanian Krishna Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut Prasad Rohit Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut Natarajan Prem Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut International Journal on Document Analysis and Recognition (IJDAR) Springer-Verlag 14/2(2011-06-01), 189-200 1433-2833 14:2<189 2011 14 10032 https://doi.org/10.1007/s10032-011-0150-z text/html Onlinezugriff via DOI 1 research-article jats NATIONALLICENCE 856 40 https://doi.org/10.1007/s10032-011-0150-z text/html Onlinezugriff via DOI NATIONALLICENCE 700 1- Subramanian Krishna Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut NATIONALLICENCE 700 1- Prasad Rohit Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut NATIONALLICENCE 700 1- Natarajan Prem Raytheon BBN Technologies, 10 Moulton Street, 02138, Cambridge, MA, USA aut NATIONALLICENCE 773 0- International Journal on Document Analysis and Recognition (IJDAR) Springer-Verlag 14/2(2011-06-01), 189-200 1433-2833 14:2<189 2011 14 10032 Metadata rights reserved Springer special CC-BY-NC licence nationallicence BK010053 XK010053 XK010000 NATIONALLICENCE NATIONALLICENCE NL-springer