This jupyter notebook is about classifying the dialogue act in a sentence. The model is trained on the SwDA and MRDA dataset.
In Natural language understanding, a dialog act is an utterance, in the context of a conversational dialog, that serves the function in the dialog. Dialog act are a type of speech act. Types of dilog acts may include question, statement or a request. The task of predicting dialog acts(DA) based on dialog system is a key factor in the development of conversational agents.
| Speaker | Utterance | DA label |
|---|---|---|
| A | Okay. | Other |
| A | Um. what did you do this weekend | Question |
| B | Well, uh, pretty much spent most of my time in this yard | Statement |
| B | [Throat Clearing] | Non verbal |
| A | Uh Huh | Backchannel |
| A | What do you have planned for your yard? | Question |
| B | Well we are in the process of, realizing it | Statement |
| Dataset | Classes | Vocabulary size | Train size | Val size | Test size |
|---|---|---|---|---|---|
| MRDA | 5 | 10k | 75k | 16k | 16k |
| SwDA | 43 | 19k | 193k | 4k | 20k |