The enclosed file annotated-corpus.txt contains 400 dialogue fragments extracted from the AMI Meeting Corpus [1], the ICSI Meeting Corpus [2] and the Switchboard Corpus [3] that have been annotated with the taxonomy presented in [4].

The individual fragments in corpus.txt are separated by double linebreaks and are annotated as follows:


The first line of a fragment starts with the string "SOURCE: " followed by the corpus the fragment was extracted from (either "AMI", "ISCI" or "SWB").


Each following line (except the last, see below) contains one utterance extracted from the source corpus formatted as follows.

- If the utterance was in its source corpus annotated as a rejection move, it starts with the string "--> ", followed by the label for the speaker of the utterance (identical to the label in the source corpus), followed by the string ": ", followed by the transcription of the utterance (as in the source corpus).

- If the utterance was in its source corpus annotated as the target of the rejection move, it starts with the string "-> ", followed by the label for the speaker of the utterance (identical to the label in the source corpus), followed by the string ": ", followed by the transcription of the utterance (as in the source corpus).

- Otherwise, it starts with the label for the speaker of the utterance (identical to the label in the source corpus), followed by the string ": ", followed by the transcription of the utterance (as in the source corpus).


The last line of a fragment contains a comma-separated list of three entries:

- First, what part of the utterance marked with "->" is being rejected by the utterance marked with "-->" (one of "content", "implicature", "precondition"; see [4] for details).

- Second, how the utterance marked with "-->" achieves its rejection force (one of "contradiction", "convers-imp", "convent-imp", "disbelief", "irony"; see [4] for details).

- Third, comments left by the annotator (may be the empty string).



[1] Jean Carletta. 2007. Unleashing the killer corpus: experiences in creating the multi-everything AMI meeting corpus. Language Resources and Evaluation, 41(2):181–190.

[2] Adam Janin, Don Baron, Jane Edwards, Dan Ellis, David Gelbart, Nelson Morgan, Barbara Peskin, Thilo Pfau, Elisabeth Shriberg, Andreas Stolcke, and Chuck Wooters. 2003. The ICSI Meeting Corpus. In Proceedings of ICASSP'03, pages 364–367.

[3] John J. Godfrey, E. C. Holliman, and J. McDaniel. 1992. SWITCHBOARD: Telephone Speech Corpus for Research and Development. IEEE Conference on Acoustics, Speech, and Signal Processing, 1:517–520.

[4] Julian J. Schlöder and Raquel Fernández. 2019. How to Reject What in Dialogue. Proceedings of the 23rd Workshop on the Semantics and Pragmatics of Dialogue (SemDial).