| Popis: |
Given the diversity of lexical-semantic resources, particularly dictionaries, integrating such resources by aligning various types of information is an important task, both in e-lexicography and natural language processing. The current study aims at analyzing the automatic alignment of word senses of the same lemmas across two comprehensive monolingual Danish dictionaries, the historic Ordbog over det Danske Sprogand the modern Den Danske Ordbog. We report our efforts in creating a gold-standard dataset and show that semantic similarity measures can be efficiently used to create statistical models to automatically align senses across dictionaries. This work has received funding from the EU’s Horizon 2020 Research and Innovation programme through the ELEXIS project under grant agreement No. 731015. The authors would like to thank Thomas Troelsgård, Society for Danish Language and Literature (DSL) who contributed to the linking of identical lemmas between ODS and DDO and Sussi Olsen, CST, University of Copenhagen who contributed to the manual annotation task. peer-reviewed |
