As of January 9, Wikipedia allows semantically enhancing mathematical formulae with the definitions of the terms they contain. For example, the formula of the famous mass-energy equivalence E=mc² now contains information on the variables E, m, and c. Clicking on the formula opens an information page (see figure) that displays the meanings of the variables and links to additional specifications of the formula’s terms.
Our group members André-Greiner Petter and Moritz Schubotz developed this new feature as part of the research for the DFG grant GI 1259-1. The additional information on formulae is obtained from Wikidata – a central, language-independent Wikimedia database that can be edited by both humans and bots. To further improve the availability of mathematical semantics, we combine the power of artificial intelligence with the capabilities of humans to (semi-automatically) generate semantic markup for mathematics in Wikipedia. Please refer to the following publications for details and join the Wikimedia Math community group if you are interested in helping to improve math support in Wikipedia.
Related publications:
Why Machines Cannot Learn Mathematics, Yet
A Greiner-Petter, T Ruas, M Schubotz, A Aizawa, W Grosky, B Gipp
Proceedings of the 4rd Joint Workshop on Bibliometric-enhanced Information Retrieval and Natural Language Processing for Digital Libraries (BIRNDL 2018) co-located with the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2019), Paris, USA, July 25, 2019.
PDF Preprint
math.wikipedia.org: A vision for a collaborative semi-formal, language independent math(s) encyclopedia
J Corneli, M Schubotz
Conference on Artificial Intelligence and Theorem Proving, 2017
PDF Bibtex
AnnoMath TeX – a formula identifier annotation recommender system for STEM documents
P Scharpf, I Mackerracher, M Schubotz, J Beel, C Breitinger, B Gipp
Proceedings of the 13th ACM Conference on Recommender Systems, RecSys 2019, Copenhagen, Denmark, September 16-20, 2019
DOI: 10.1145/3298689.3347042 Preprint Bibtex Homepage Core Rank B
