77 Lexicography in gLobaL contexts The DHmine Dictionary Work-fow: Creating a Knowledge-based Author’s Dictionary Tamás Mészáros 1 , Margit Kiss 2 1 Budapest University of Technology and Economics, 2 Institute for Literary Studies, Hungarian Academy of Sciences E-mail: meszaros@mit.bme.hu, kiss.margit@btk.mta.hu Abstract Digitalized author’s dictionaries could play an important role in humanities research. Not only could they pro- vide better ways to study an individual author’s vocabulary, but they could also act as a knowledge source for other computer-based methods. We present the process of making an author’s dictionary of headwords, writ- ing variations, word forms and corpus citations extended with part-of-speech, linguistic, literary and semantic information. We also describe how this extended dictionary incorporates knowledge from linked open data sources and from critical annotations and builds an RDF knowledge base attached to the dictionary. The result is a vast knowledge source about an author’s oeuvre that can be studied and used to enhance corpus analysis. We demonstrate our method on processing a large text corpora of 1.5 million words from the 18th century and on creating the digital author’s dictionary of Kelemen Mikes. Keywords: author’s dictionaries, knowledge-based systems, corpus analysis, linked open data 1 Aim of the Research The ongoing DHmine project at the Budapest University of Technology and Economics aims to create a software tools to support various digital humanities (DH) research tasks (Mészáros 2016). In cooperation with the Institute for Literary Studies of Hungarian Academy of Sciences, we pro- cessed the works of Kelemen Mikes, an 18th-century author often called the “Hungarian Goethe” (Kiss 2012). Our main goal was to create an author’s dictionary of Kelemen Mikes. This was a groundbreaking work since this era is rather underrepresented in computerized corpus building, and no complete dig- ital author’s dictionary had been created in Hungarian language before. Our aim was thus to establish a work-fow for creating such dictionaries and also to demonstrate the possible benefts of informa- tion technology in this feld. We concentrated our work on two main aspects: increasing the efciency of the dictionary-making process by utilizing various software tools, and taking a step beyond data-centric digitalization and introducing knowledge-based methods in creating and using the dictionary. Our research aim was to develop methods for incorporating various kinds of knowledge in digital author’s dictionaries and then utilize them in corpus analysis. 2 Previous Work Computerized tools play an increasingly important role in humanities research. They provide efcient tools for storing, searching, retrieving and displaying digitalized texts, they also vastly improve the 1 / 10