chapter  12
Lexicography and natural language processing
ByAleš Horák, Adam Rambousek
Pages 18

The chapter introduces links between lexicography and natural language processing in both directions. First, how the lexicographic work benefits from techniques of natural language processing such as word sense disambiguation, semantic similarity measures, named entity recognition, relation extraction, or corpora good example search. Second, intensive exploitation of lexicographic data in language engineering technologies, in the form of general dictionaries or encyclopaedias as well as specialized lexicons, ontologies, or lexicographic databases. The chapter concentrates on selected topics of natural language processing (NLP) that are often advantageously used in lexicographic processes. During the 1990s dictionary publishers started to buy or develop in-house specialized dictionary writing systems, and lexicographers were moving from papers to computers. In the field of lexicography, the most prominent crowdsourced resource is the Wiktionary, a sister project of Wikipedia. The goal of Wiktionary is to create a freely available "dictionary of all words in all languages" edited by volunteers.