QTLeap Corpus V1.2
|Handle:||https://hdl.handle.net/21.11129/0000-000B-D36B-F (persistent URL to this page)|
The QTLeap corpus is composed by 4000 question and answer pairs in the domain of computer and IT troubleshooting for both hardware and software. This material was collected using a support service via chat, this implies that the corpus is composed by naturally occurring utterances produced by users while interacting with a service.
The corpus was collected selecting the data contained in a database where all the interactions with the clients are saved. Only interactions composed by one question and the respective answer were included in the corpus.