Towards machine-actionable modules of a digital mathematics library: the example of DML-CZ

Authors:
Michal Růžička;Petr Sojka;Vlastimil Krejčíř
Affiliations:
Masaryk University, Faculty of Informatics, Brno, Czech Republic and Masaryk University, Institute of Computer Science, Brno, Czech Republic;Masaryk University, Faculty of Informatics, Brno, Czech Republic;Masaryk University, Faculty of Informatics, Brno, Czech Republic and Masaryk University, Institute of Computer Science, Brno, Czech Republic
Venue:
CICM'13 Proceedings of the 2013 international conference on Intelligent Computer Mathematics
Year:
2013

Citing 3
Cited 0

INFTY: an integrated OCR system for mathematical documents

Proceedings of the 2003 ACM symposium on Document engineering
The art of mathematics retrieval

Proceedings of the 11th ACM symposium on Document engineering
Logical Structure Recovery in Scholarly Articles with Rich Document Features

International Journal of Digital Library Systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

Publishing and archiving mathematical literature presents its own sets of problems. Reaching the goal of building global digital mathematics library (DML), smaller DMLs play an inevitable role in collecting, validating, digitizing and checking data from smaller publishers. In this paper, we overview the technical challenges of building a machine-actionable set of modules we have developed over almost a decade of evolution of the Czech Digital Mathematics Library (DML-CZ). Firstly, we survey methods of effective automated data acquisition from the content providers. Then we show OCR processing of mathematical documents and automated segmentation of plain text references for metadata enhancement and effective DOI look up. Finally we describe connection to the European Digital Mathematics Library (EuDML) project and public interfaces of DML-CZ for the best visibility and accessibility.