Automated Classification and Categorization of Mathematical Knowledge

Warning

This publication doesn't include Faculty of Education. It includes Faculty of Informatics. Official publication website can be found on muni.cz.

Authors	ŘEHŮŘEK Radim SOJKA Petr
Year of publication	2008
Type	Article in Proceedings
Conference	Intelligent Computer Mathematics: AISC/Calculemus/MKM LNAI 5144
MU Faculty or unit	Faculty of Informatics
Citation
Web	web of conference ACM Portal paper preprint Springer Link
Field	Informatics
Keywords	machine learning; classification; categorization; similarity of mathematical papers; mathematical knowledge management; MSC;mathematical subject classification
Description	There is a common Mathematics Subject Classification (MSC) System used for categorizing mathematical papers and knowledge. We present results of machine learning of the MSC on full texts of papers in the mathematical digital libraries DML-CZ and NUMDAM. The F1-measure achieved on classification task of top-level MSC categories exceeds 89%. We describe and evaluate our methods for measuring the similarity. of papers in the digital library based on paper full texts.
Related projects:	DML-CZ: Czech Digital Mathematical Library E-learning in the Semantic Web Context Centrum komputační lingvistiky Prostředky tvorby komplexní báze znalostí pro komunikaci se sémantickým webem v přirozeném jazyce