Finland Swedish Text Corpus (FISC)

Department of Scandinavian languages and literature
P.O.Box 4
FIN-00014 University of Helsinki

Personnel

• Mirja Saari (Project Director)
• Jan Lindstrom (Principal Researcher)
• Juhani Birn (Researcher 1994)
• Hannu Aarnitukia (Researcher 1991-1993)
• Eva Hannus (Researcher Assistant 1991-1992)
• Maria Hagglund (Researcher Assistant 1995)
• Annika Salervo (Researcher Assistant 1993-1994)

Key points

Size:

• The corpus contains about 2.5 million running words of modern written Swedish texts published in Finland (1990s). The kernel corpus is accompanied by a minor section on spoken language.

• Following text types are included and structured as separate sections of the corpus:
  • newspapers
  • literature (fiction)
  • non-literary prose (non-fiction)
  • official documents (legal texts, administration)
  • casual conversations

• The texts in the corpus contain embedded tags indicating bibliograhic details, headings, captions, paragrahps and other significant textual features. The tags conform to TEI-recommendations.

• The corpus is located at the Department of Linguistic at the University of Helsinki in a language corpus server (the UHLCS-server) running with a UNIX operating system.

• The language corpus server offers tools for processing the data and search facilities including concordance listings (e.g. KWIC) and a morphological analyzer for Swedish texts (SWETWOL). Coventional unix-commands are, of course, available as well.

• If you have access to Internet you are in practise able to contact the language corpus server, be your personal terminal a PC, a Macintosh, or a UNIX-machine. Preferred protocol is Telnet.

• Anyone who wishes access to the corpus is requested to contact the project personnel. Users need a personal password to enter the University of Helsinki Language Corpus Server.

Uses of the corpus

Linguistic research:

• The corpus will provide a database for a systematic mapping of modern Finland Swedish.

Language guidance:

• The corpus will offer a body of natural linguistic data for advisory purposes concerning grammar and usage.

Lexicography:

• The corpus can be used as empirical reference material in lexicographic work.

Language contact:

• The corpus may offer information on language contact phenomena between Finnish and Swedish in Finland.

Language technology:

• The corpus has been used as a basis for the development of software for linguistic analysis (e.g. the SWETWOL-analyzer).

The consortium

The project has been conducted by a consortium comprising:

• The Department of Scandinavian Languages (University of Helsinki)
• The Department of General Linguistics (University of Helsinki)
• The Finnish Research Centre for Domestic Languages

The Department of Scandinavian Languages (Nordica) is the Lead Participant of the consortium.

The project started officially at Nordica in August 1991. The kernel corpus of about 2,5 running words was established by the end of 1995.

Future plans

• The kernel corpus of 2,5 million words may be complemented with a monitor corpus of no finite size.
• There is good possibility to extend the section on spoken language. Today, there is a material of about 80.000 running word consisting of transcripts of speech. The material was originally collected and transcribed in the project Swedish Coversations in Helsinki (SAM).
• Also a historic reference corpus (1600s, 1700s, 1800s) is beeing planned. There is a material of about 200.000 words of 18th century Finland Swedish but the processing of these texts is yet at a preliminary stage.