Extending the CzeSL corpus – Instytut Slawistyki Zachodniej i Południowej

1. Instructing Polish students about collecting texts, explaining the purpose of the project

2. Filling in metadata questionnaires

Person ID, e.g. TOU_H305
Gender
Age
Age category: 6-11; 12-15; 16
First language: two-character code according to ISO 639-1
First language group: Indo-European (IE), non-Indo-European (nIE), Slavic (S)
Knowledge of other foreign languages: ISO code
Proficiency in Czech at the time of writing: A1; A1+; A2; A2+; B1; B2; C1; C2
Knowledge of Czech in the family: mother; father; partner; siblings; others, nobody
Length of stay in Czechia in years: –1; 1; 2; 2–
Completed or current Czech language courses: individual, commercial, self-taught, university, abroad, primary school; secondary school, other
Intensity of Czech language tuition in hours per week: –3; 5–15; 15–
Textbooks used: Basic Czech (BC), Communicative Czech (CC), Čeština pro economy (CE), Chcete mluvit česky? (CMC), Čeština pro cizince (CpC), Easy Czech Elementary (ECE), New Czech Step by Step (NCSS), other
bilingual: yes; no

b. About the text

Text ID, e.g. TOU_H305_442
Date when the text was collected (YYYY-MM-DD)
Medium of the text: manuscript; PC
Time limit for writing the text in minutes: 10; 15; 20; 30; 40; 45; 60; other; no
Additional help during writing the text: dictionary; student’s book; other; no
Was the text written during an exam: interim; final; no
Limit in words
Title of the text, e.g. The event that changed my life
Topic type: general; specific
Activity before writing the text: practice; discussion; visual; vocabulary; other; no
Ability to choose the topic: selection from many; assigned topic; any; other
Genre: any; assigned
Actual text type: informative; descriptive; opinion; short story
Actual number of words

3. Collecting students’ essays

4. Digitization of files

5. Release of an extended version of CzeSL-SGT with automatic annotation in KonText (the Czech team):

1. Automatic error annotation (suggested corrections from a spell/grammar checker, error type identifier)
2. Automatic linguistic annotation (tags and lemmas for the original and the corrected form)
3. Adding metadata

6. Preparation of tasks for Polish students of Czech based on CzeSL-SGT (the Polish team, work in progress)

7. Release of an extended version of CzeSL with manual error annotation in TEITOK and KonText (the Czech team, to do)

1. Manual multi-level correction in TEITOK, using existing manual annotation in feat (the Czech team, work in progress)
2. Automatic linguistic annotation