Datasets:

Languages:
English
Multilinguality:
monolingual
Size Categories:
10K<n<100K
Language Creators:
crowdsourced
Annotations Creators:
expert-generated
Source Datasets:
original
Tags:
License:
ArnavL commited on
Commit
dd5a3d3
1 Parent(s): f5d7cb8

Added dataset information in clinic oos dataset card (#4751)

Browse files

* Added Dataset information in Clinic oos card

* Added Field and Instance Information

* Added Label List in Data Fields

* Updated Table Caption

* Update datasets/clinc_oos/README.md

* Update README.md

Co-authored-by: Quentin Lhoest <42851186+lhoestq@users.noreply.github.com>

Commit from https://github.com/huggingface/datasets/commit/f9713d2e23813142a02f1b0e965095f528785cff

Files changed (1) hide show
  1. README.md +203 -10
README.md CHANGED
@@ -21,7 +21,7 @@ paperswithcode_id: clinc150
21
  pretty_name: CLINC150
22
  ---
23
 
24
- # Dataset Card for [Dataset Name]
25
 
26
  ## Table of Contents
27
  - [Dataset Description](#dataset-description)
@@ -52,34 +52,210 @@ pretty_name: CLINC150
52
  - **Homepage:** [Github](https://github.com/clinc/oos-eval/)
53
  - **Repository:** [Github](https://github.com/clinc/oos-eval/)
54
  - **Paper:** [Aclweb](https://www.aclweb.org/anthology/D19-1131)
55
- - **Leaderboard:**
56
  - **Point of Contact:**
57
 
58
  ### Dataset Summary
59
 
60
- [More Information Needed]
61
 
62
  ### Supported Tasks and Leaderboards
63
 
64
- [More Information Needed]
65
 
66
  ### Languages
67
 
68
- [More Information Needed]
69
 
70
  ## Dataset Structure
71
 
72
  ### Data Instances
73
 
74
- [More Information Needed]
 
 
 
 
 
 
75
 
76
  ### Data Fields
77
 
78
- [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
79
 
80
  ### Data Splits
81
 
82
- [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
83
 
84
  ## Dataset Creation
85
 
@@ -136,8 +312,25 @@ pretty_name: CLINC150
136
  [More Information Needed]
137
 
138
  ### Citation Information
139
-
140
- [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
141
  ### Contributions
142
 
143
  Thanks to [@sumanthd17](https://github.com/sumanthd17) for adding this dataset.
 
21
  pretty_name: CLINC150
22
  ---
23
 
24
+ # Dataset Card for CLINC150
25
 
26
  ## Table of Contents
27
  - [Dataset Description](#dataset-description)
 
52
  - **Homepage:** [Github](https://github.com/clinc/oos-eval/)
53
  - **Repository:** [Github](https://github.com/clinc/oos-eval/)
54
  - **Paper:** [Aclweb](https://www.aclweb.org/anthology/D19-1131)
55
+ - **Leaderboard:** [PapersWithCode](https://paperswithcode.com/sota/text-classification-on-clinc-oos)
56
  - **Point of Contact:**
57
 
58
  ### Dataset Summary
59
 
60
+ Task-oriented dialog systems need to know when a query falls outside their range of supported intents, but current text classification corpora only define label sets that cover every example. We introduce a new dataset that includes queries that are out-of-scope (OOS), i.e., queries that do not fall into any of the system's supported intents. This poses a new challenge because models cannot assume that every query at inference time belongs to a system-supported intent class. Our dataset also covers 150 intent classes over 10 domains, capturing the breadth that a production task-oriented agent must handle. It offers a way of more rigorously and realistically benchmarking text classification in task-driven dialog systems.
61
 
62
  ### Supported Tasks and Leaderboards
63
 
64
+ - `intent-classification`: This dataset is for evaluating the performance of intent classification systems in the presence of "out-of-scope" queries, i.e., queries that do not fall into any of the system-supported intent classes. The dataset includes both in-scope and out-of-scope data. [here](https://paperswithcode.com/sota/text-classification-on-clinc-oos).
65
 
66
  ### Languages
67
 
68
+ English
69
 
70
  ## Dataset Structure
71
 
72
  ### Data Instances
73
 
74
+ A sample from the training set is provided below:
75
+ ```
76
+ {
77
+ 'text' : 'can you walk me through setting up direct deposits to my bank of internet savings account',
78
+ 'label' : 108
79
+ }
80
+ ```
81
 
82
  ### Data Fields
83
 
84
+ - text : Textual data
85
+ - label : 150 intent classes over 10 domains, the dataset contains one label for 'out-of-scope' intent.
86
+
87
+ The Label Id to Label Name map is mentioned in the table below:
88
+
89
+ | **Label Id** | **Label name** |
90
+ |--- |--- |
91
+ | 0 | restaurant_reviews |
92
+ | 1 | nutrition_info |
93
+ | 2 | account_blocked |
94
+ | 3 | oil_change_how |
95
+ | 4 | time |
96
+ | 5 | weather |
97
+ | 6 | redeem_rewards |
98
+ | 7 | interest_rate |
99
+ | 8 | gas_type |
100
+ | 9 | accept_reservations |
101
+ | 10 | smart_home |
102
+ | 11 | user_name |
103
+ | 12 | report_lost_card |
104
+ | 13 | repeat |
105
+ | 14 | whisper_mode |
106
+ | 15 | what_are_your_hobbies |
107
+ | 16 | order |
108
+ | 17 | jump_start |
109
+ | 18 | schedule_meeting |
110
+ | 19 | meeting_schedule |
111
+ | 20 | freeze_account |
112
+ | 21 | what_song |
113
+ | 22 | meaning_of_life |
114
+ | 23 | restaurant_reservation |
115
+ | 24 | traffic |
116
+ | 25 | make_call |
117
+ | 26 | text |
118
+ | 27 | bill_balance |
119
+ | 28 | improve_credit_score |
120
+ | 29 | change_language |
121
+ | 30 | no |
122
+ | 31 | measurement_conversion |
123
+ | 32 | timer |
124
+ | 33 | flip_coin |
125
+ | 34 | do_you_have_pets |
126
+ | 35 | balance |
127
+ | 36 | tell_joke |
128
+ | 37 | last_maintenance |
129
+ | 38 | exchange_rate |
130
+ | 39 | uber |
131
+ | 40 | car_rental |
132
+ | 41 | credit_limit |
133
+ | 42 | oos |
134
+ | 43 | shopping_list |
135
+ | 44 | expiration_date |
136
+ | 45 | routing |
137
+ | 46 | meal_suggestion |
138
+ | 47 | tire_change |
139
+ | 48 | todo_list |
140
+ | 49 | card_declined |
141
+ | 50 | rewards_balance |
142
+ | 51 | change_accent |
143
+ | 52 | vaccines |
144
+ | 53 | reminder_update |
145
+ | 54 | food_last |
146
+ | 55 | change_ai_name |
147
+ | 56 | bill_due |
148
+ | 57 | who_do_you_work_for |
149
+ | 58 | share_location |
150
+ | 59 | international_visa |
151
+ | 60 | calendar |
152
+ | 61 | translate |
153
+ | 62 | carry_on |
154
+ | 63 | book_flight |
155
+ | 64 | insurance_change |
156
+ | 65 | todo_list_update |
157
+ | 66 | timezone |
158
+ | 67 | cancel_reservation |
159
+ | 68 | transactions |
160
+ | 69 | credit_score |
161
+ | 70 | report_fraud |
162
+ | 71 | spending_history |
163
+ | 72 | directions |
164
+ | 73 | spelling |
165
+ | 74 | insurance |
166
+ | 75 | what_is_your_name |
167
+ | 76 | reminder |
168
+ | 77 | where_are_you_from |
169
+ | 78 | distance |
170
+ | 79 | payday |
171
+ | 80 | flight_status |
172
+ | 81 | find_phone |
173
+ | 82 | greeting |
174
+ | 83 | alarm |
175
+ | 84 | order_status |
176
+ | 85 | confirm_reservation |
177
+ | 86 | cook_time |
178
+ | 87 | damaged_card |
179
+ | 88 | reset_settings |
180
+ | 89 | pin_change |
181
+ | 90 | replacement_card_duration |
182
+ | 91 | new_card |
183
+ | 92 | roll_dice |
184
+ | 93 | income |
185
+ | 94 | taxes |
186
+ | 95 | date |
187
+ | 96 | who_made_you |
188
+ | 97 | pto_request |
189
+ | 98 | tire_pressure |
190
+ | 99 | how_old_are_you |
191
+ | 100 | rollover_401k |
192
+ | 101 | pto_request_status |
193
+ | 102 | how_busy |
194
+ | 103 | application_status |
195
+ | 104 | recipe |
196
+ | 105 | calendar_update |
197
+ | 106 | play_music |
198
+ | 107 | yes |
199
+ | 108 | direct_deposit |
200
+ | 109 | credit_limit_change |
201
+ | 110 | gas |
202
+ | 111 | pay_bill |
203
+ | 112 | ingredients_list |
204
+ | 113 | lost_luggage |
205
+ | 114 | goodbye |
206
+ | 115 | what_can_i_ask_you |
207
+ | 116 | book_hotel |
208
+ | 117 | are_you_a_bot |
209
+ | 118 | next_song |
210
+ | 119 | change_speed |
211
+ | 120 | plug_type |
212
+ | 121 | maybe |
213
+ | 122 | w2 |
214
+ | 123 | oil_change_when |
215
+ | 124 | thank_you |
216
+ | 125 | shopping_list_update |
217
+ | 126 | pto_balance |
218
+ | 127 | order_checks |
219
+ | 128 | travel_alert |
220
+ | 129 | fun_fact |
221
+ | 130 | sync_device |
222
+ | 131 | schedule_maintenance |
223
+ | 132 | apr |
224
+ | 133 | transfer |
225
+ | 134 | ingredient_substitution |
226
+ | 135 | calories |
227
+ | 136 | current_location |
228
+ | 137 | international_fees |
229
+ | 138 | calculator |
230
+ | 139 | definition |
231
+ | 140 | next_holiday |
232
+ | 141 | update_playlist |
233
+ | 142 | mpg |
234
+ | 143 | min_payment |
235
+ | 144 | change_user_name |
236
+ | 145 | restaurant_suggestion |
237
+ | 146 | travel_notification |
238
+ | 147 | cancel |
239
+ | 148 | pto_used |
240
+ | 149 | travel_suggestion |
241
+ | 150 | change_volume |
242
 
243
  ### Data Splits
244
 
245
+ The dataset comes in different subsets:
246
+
247
+ - `small` : Small, in which there are only 50 training queries per each in-scope intent
248
+ - `imbalanced` : Imbalanced, in which intents have either 25, 50, 75, or 100 training queries.
249
+ - `plus`: OOS+, in which there are 250 out-of-scope training examples, rather than 100.
250
+
251
+
252
+ | name |train|validation|test|
253
+ |----------|----:|---------:|---:|
254
+ |small|7600| 3100| 5500 |
255
+ |imbalanced|10625| 3100| 5500|
256
+ |plus|15250| 3100| 5500|
257
+
258
+
259
 
260
  ## Dataset Creation
261
 
 
312
  [More Information Needed]
313
 
314
  ### Citation Information
315
+ ```
316
+ @inproceedings{larson-etal-2019-evaluation,
317
+ title = "An Evaluation Dataset for Intent Classification and Out-of-Scope Prediction",
318
+ author = "Larson, Stefan and
319
+ Mahendran, Anish and
320
+ Peper, Joseph J. and
321
+ Clarke, Christopher and
322
+ Lee, Andrew and
323
+ Hill, Parker and
324
+ Kummerfeld, Jonathan K. and
325
+ Leach, Kevin and
326
+ Laurenzano, Michael A. and
327
+ Tang, Lingjia and
328
+ Mars, Jason",
329
+ booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)",
330
+ year = "2019",
331
+ url = "https://www.aclweb.org/anthology/D19-1131"
332
+ }
333
+ ```
334
  ### Contributions
335
 
336
  Thanks to [@sumanthd17](https://github.com/sumanthd17) for adding this dataset.