2000 Communicator Dialogue Act Tagged
|Item Name:||2000 Communicator Dialogue Act Tagged|
|Author(s):||Rashmi Prasad, Marilyn Walker|
|LDC Catalog No.:||LDC2004T15|
|Release Date:||June 15, 2004|
|Data Source(s):||telephone conversations|
|Application(s):||nominal expression generation, speech recognition, spoken dialogue modeling, spoken dialogue systems, summarization, tagging, topic detection and tracking|
LDC User Agreement for Non-Members
|Online Documentation:||LDC2004T15 Documents|
|Licensing Instructions:||Subscription & Standard Members, and Non-Members|
|Citation:||Prasad, Rashmi, and Marilyn Walker. 2000 Communicator Dialogue Act Tagged LDC2004T15. Web Download. Philadelphia: Linguistic Data Consortium, 2004.|
2000 Communicator Dialogue Act Tagged was developed by the Linguistic Data Consortium (LDC) and contains approximately 314,000 words of system and user interactions with entity and dialogue act tagging.
This release is an addendum to 2000 Communicator Evaluation (LDC2002S56) developed by LDC in 2002. This addendum contains annotations on the transcriptions of the system and user utterances as taken from the log files of LDC2002S56.
Dialogue Act annotations are provided for system utterances in the dialogues. The dialogue act tags follow the DATE (Dialogue Act Tagging for Evaluation) scheme. In addition, both system and user utterances are tagged for named entities. For further info on the 2000 Communicator Evaluation corpus, please refer to the main publication from 2002 linked above.
The complete Dialogue Act annotated corpus is available as a single XML text file totalling approximately 16 MB.
Dialogue Act tagging was done automatically via pattern matching with human-labeled dialogue utterances used by the nine different participating Communicator Systems. Named entity tagging also followed the same methodology. Here is the breakdown for dialogues and dialogue acts:
|Dialogues||Dialogue Acts||Tagged Dialogue Acts||Unique Tags|
Each dialogue is segmented into system and user turns. Except for one system, no utterance segmentation was done within the turns in the log files. The number of utterances is therefore the same as the number of turns. Utterance segmentation is carried out and reflected as the dialogue act segmentation. Here is a breakdown of the distribution of turns, and words:
The release also includes the raw transcripts from the dialogues.
For an example of the data in this corpus, please view this sample (TXT).
This research was conducted using funding from the following grant number and funding agency: DARPA - contract MDA972-99-3-0003.
None at this time.