Preface Data Documentation Disclaimer

Preface

The Croatian Valency Lexicon of Verbs, Version 2.0008 (CROVALLEX 2.0008) is an attempt of formal description of valency frames of Croatian verbs. CROVALLEX 2.0008 was developed as the part of the PhD thesis titled Approaches to the Development of the Machine Lexicon for Croatian Language written by Nives Mikelic Preradovic and supervised by prof.dr.sc. Damir Boras at the Department of Information Sciences, Faculty of Humanistics and Social Sciences, Zagreb University.

The Functional Generative Description (FGD), being developed by Czech linguists Petr Sgall and his collaborators since the 1960s, is used as the background theory in CROVALLEX 2.0008. for the description of valency frames of selected verbs.

CROVALLEX 2.0008 contains roughly 1740 verbs. They were selected from the Croatian frequency dictionary, according to their number of occurrences.

The preparation of this version of CROVALLEX has taken around three years and is still a work in progress requiring further linguistic research.

Data

The logical structure of the CROVALLEX data is described here. The structure is based on the Valency Lexicon of Czech Verbs (VALLEX 1.0) developed by Zdeněk Žabokrtský and Markéta Lopatková at the Institute of Formal and Applied Linguistics (ÚFAL), Faculty of Mathematics and Physics, Charles University in Prague 2004.

If you are facing problems viewing any of the accents of the verbs, please download the appropriate font here.

The lexicon is available in the following formats:

• Browsable version. HTML version of the data allows intuitive navigation through the lexicon. Verbs and frames are organized in several ways, following various criteria. You can start with the alphabetically sorted verbs.
  
  
• XML version. You can run sophisticated queries on the data (e.g. using the XSH environment, or use the data in your applications.
• Document Type Definition data/xml/CROVALLEX.dtd
• XML Data data/xml/CROVALLEX.xml

XML is the primary data format of CROVALLEX; both printable and browsable versions were automatically generated from the XML.

Documentation

The logical structure of CROVALLEX gives a short description of the lexicon. The more detailed description is available as the part of the PhD thesis titled Approaches to the Development of the Machine Lexicon for Croatian Language written by Nives Mikelic Preradovic.

Disclaimer

Although the large efforts were spent on removing annotation errors, some of them still remain in the lexicon. If you find any, you are kindly asked to report them to nmikelic@ffzg.hr. Any other comments are welcome as well.