Transformation of Wiktionary entry structure into tables and relations in a relational database schema

A. A. Krizhanovsky

DOI: https://doi.org/10.48550/arXiv.1011.1368

2010-11-05

Information Retrieval

Abstract:This paper addresses the question of automatic data extraction from the Wiktionary, which is a multilingual and multifunctional dictionary. Wiktionary is a collaborative project working on the same principles as the Wikipedia. The Wiktionary entry is a plain text from the text processing point of view. Wiktionary guidelines prescribe the entry layout and rules, which should be followed by editors of the dictionary. The presence of the structure of a Wiktionary article and formatting rules allows transforming the Wiktionary entry structure into tables and relations in a relational database schema, which is a part of a machine-readable dictionary (MRD). The paper describes how the flat text of the Wiktionary entry was extracted, converted, and stored in the specially designed relational database. The MRD contains the definitions, semantic relations, and translations extracted from the English and Russian Wiktionaries. The parser software is released under the open source license agreement (GPL), to facilitate its dissemination, modification and upgrades, to draw researchers and programmers into parsing other Wiktionaries, not only Russian and English.