Upload README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,68 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Dataset Card for WOZ 2.0
|
| 2 |
+
|
| 3 |
+
- **Repository:** https://github.com/nmrksic/neural-belief-tracker/tree/master/data/woz
|
| 4 |
+
- **Paper:** https://aclanthology.org/P17-1163.pdf
|
| 5 |
+
- **Leaderboard:** None
|
| 6 |
+
- **Who transforms the dataset:** Qi Zhu(zhuq96 at gmail dot com)
|
| 7 |
+
|
| 8 |
+
### Dataset Summary
|
| 9 |
+
|
| 10 |
+
Describe the dataset.
|
| 11 |
+
|
| 12 |
+
- **How to get the transformed data from original data:**
|
| 13 |
+
- download `woz_[train|validate|test]_en.json` from https://github.com/nmrksic/neural-belief-tracker/tree/master/data/woz and save to `woz` dir in the current directory.
|
| 14 |
+
- Run `python preprocess.py` in the current directory.
|
| 15 |
+
- **Main changes of the transformation:**
|
| 16 |
+
- domain is set to **restaurant**.
|
| 17 |
+
- normalize the value of categorical slots in state and dialogue acts.
|
| 18 |
+
- `belief_states` in WOZ dataset contains `request` intents, which are ignored in processing.
|
| 19 |
+
- use simple string match to find value spans of non-categorical slots.
|
| 20 |
+
|
| 21 |
+
- **Annotations:**
|
| 22 |
+
- User dialogue acts, state
|
| 23 |
+
|
| 24 |
+
### Supported Tasks and Leaderboards
|
| 25 |
+
|
| 26 |
+
NLU, DST, E2E
|
| 27 |
+
|
| 28 |
+
### Languages
|
| 29 |
+
|
| 30 |
+
English
|
| 31 |
+
|
| 32 |
+
### Data Splits
|
| 33 |
+
|
| 34 |
+
| split | dialogues | utterances | avg_utt | avg_tokens | avg_domains | cat slot match(state) | cat slot match(goal) | cat slot match(dialogue act) | non-cat slot span(dialogue act) |
|
| 35 |
+
|------------|-------------|--------------|-----------|--------------|---------------|-------------------------|------------------------|--------------------------------|-----------------------------------|
|
| 36 |
+
| train | 600 | 4472 | 7.45 | 11.37 | 1 | 100 | - | 100 | 96.56 |
|
| 37 |
+
| validation | 200 | 1460 | 7.3 | 11.28 | 1 | 100 | - | 100 | 95.52 |
|
| 38 |
+
| test | 400 | 2892 | 7.23 | 11.49 | 1 | 100 | - | 100 | 94.83 |
|
| 39 |
+
| all | 1200 | 8824 | 7.35 | 11.39 | 1 | 100 | - | 100 | 95.83 |
|
| 40 |
+
|
| 41 |
+
1 domains: ['restaurant']
|
| 42 |
+
- **cat slot match**: how many values of categorical slots are in the possible values of ontology in percentage.
|
| 43 |
+
- **non-cat slot span**: how many values of non-categorical slots have span annotation in percentage.
|
| 44 |
+
|
| 45 |
+
### Citation
|
| 46 |
+
|
| 47 |
+
```
|
| 48 |
+
@inproceedings{mrksic-etal-2017-neural,
|
| 49 |
+
title = "Neural Belief Tracker: Data-Driven Dialogue State Tracking",
|
| 50 |
+
author = "Mrk{\v{s}}i{\'c}, Nikola and
|
| 51 |
+
{\'O} S{\'e}aghdha, Diarmuid and
|
| 52 |
+
Wen, Tsung-Hsien and
|
| 53 |
+
Thomson, Blaise and
|
| 54 |
+
Young, Steve",
|
| 55 |
+
booktitle = "Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
|
| 56 |
+
month = jul,
|
| 57 |
+
year = "2017",
|
| 58 |
+
address = "Vancouver, Canada",
|
| 59 |
+
publisher = "Association for Computational Linguistics",
|
| 60 |
+
url = "https://aclanthology.org/P17-1163",
|
| 61 |
+
doi = "10.18653/v1/P17-1163",
|
| 62 |
+
pages = "1777--1788",
|
| 63 |
+
}
|
| 64 |
+
```
|
| 65 |
+
|
| 66 |
+
### Licensing Information
|
| 67 |
+
|
| 68 |
+
Apache License, Version 2.0
|