German Subcategorisation Lexicon

The counts in this lexicon are built using our subcategorisation frame (SCF) tagger described in the paper above; the tagger operates here using dependency parses of the SdeWaC corpus, as delivered by the most recent version of the mate-tools parser (instead of the Berkeley Parser + edge labeller combination introduced in our paper). Verb instances are tagged with a single SCF code derived from:

Schulte im Walde, S. (2002). A subcategorisation lexicon for German verbs induced from a lexicalised PCFG. In Proceedings of the 3rd Conference on Language Resources and Evaluation (LREC) (pp. 1351–1357).

Other syntactic information in the SCF database follows the annotation scheme used in the TIGER corpus.

For each verb in SdeWaC, the SCF database contains a single line entry. This entry is in tab-separated format; the columns contain the following information:

1. The index of the sentence in SdeWaC (an integer from 0 to 45400445);
2. The index of the verb in the current sentence;
3. The STTS part of speech tag of the verb;
4. The lemmatised form of the verb;
5. The subcategorisation code of the verb;
6. A Boolean field containing True if the verb is in the passive voice and False otherwise;
7. A code indicating the clause type of phrase containing the verb;
8. The index of the verb's governor in the sentence, if found: a governor can be a modal verb, an auxiliary verb for tense constructions, or an aspectual verb for certain kinds of aspectual constructions;
9. The lemmatised form of the governing verb, if found.

Fields 10 and up list information pertaining to the verb's complements in the sentence; the total number of fields varies with the number of complements to the verb. There are eleven fields per complement:

1. The index in the sentence of the complement's head;
2. The simplified part of speech of the complement (one of n, v, or p);
3. The full STTS part of speech code of the complement's head;
4. The complement's head;
5. The lemmatised form of the complement's head;
6. The edge label of the complement;
7. The index in the sentence of the complement's argument (only if the complement is prepositional);
8. The part of speech of the complement's argument;
9. The complement's argument;
10. The lemmatised form of the complement's argument;
11. Morphological analysis of the complement's argument.