yamasy1549 commited on
Commit
a5be0f5
1 Parent(s): be73f3d

:memo: Add README

Browse files
Files changed (1) hide show
  1. README.md +258 -0
README.md ADDED
@@ -0,0 +1,258 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ language:
3
+ - ja
4
+ license: cc-by-sa-4.0
5
+ tags:
6
+ - nlp
7
+ - japanese
8
+ - dialogue
9
+ - dialogue-corpus
10
+ - dialogue-system
11
+ language_creators:
12
+ - crowdsourced
13
+ pretty_name: RealPersonaChat
14
+ size_categories:
15
+ - 10K<n<100K
16
+ task_categories:
17
+ - text-generation
18
+ - text-classification
19
+ task_ids:
20
+ - dialogue-modeling
21
+ - dialogue-generation
22
+ multilinguality:
23
+ - monolingual
24
+ source_datasets:
25
+ - original
26
+ ---
27
+
28
+
29
+ [![CC BY-SA 4.0](https://img.shields.io/badge/License-CC%20BY%20SA%204.0-green.svg)](https://creativecommons.org/licenses/by-sa/4.0/deed.ja)
30
+ [![RealPersonaChatのバージョン](https://img.shields.io/github/v/release/nu-dialogue/real-persona-chat)](https://github.com/nu-dialogue/real-persona-chat)
31
+
32
+
33
+ # Dataset Card for RealPersonaChat
34
+
35
+ ## Dataset Description
36
+
37
+ - **Homepage:** [nu-dialogue/real-persona-chat](https://github.com/nu-dialogue/real-persona-chat)
38
+ - **Repository:** [nu-dialogue/real-persona-chat](https://github.com/nu-dialogue/real-persona-chat)
39
+ - **Paper:** [RealPersonaChat: A realistic persona chat corpus with interlocutors' own personalities](https://aclanthology.org/2023.paclic-1.85)
40
+
41
+
42
+ ### Dataset Summary
43
+
44
+ RealPersonaChat は,話者本人のペルソナと性格特性を含む,約14,000件の日本語雑談対話からなるコーパスです.詳細は [GitHub](https://github.com/nu-dialogue/real-persona-chat) をご覧ください.
45
+
46
+ > [!CAUTION]
47
+ > **本コーパスの使用にあたっては,次のことに十分注意してください.**
48
+ > * 本コーパスのデータから個人を特定しようとしないこと.
49
+ > * 本コーパスを,特定の話者へのなりすましに用いないこと.
50
+ > * 本コーパスを話者の属性や性格特性の推定などに用いる際は,自身の情報を推定されたくない話者の権利についても留意すること [(Tatman, 2020)](https://www.rctatman.com/files/Tatman_2020_WiNLP_Keynote.pdf).
51
+
52
+
53
+ ## Dataset Usage
54
+
55
+ ```python
56
+ from datasets import load_dataset
57
+
58
+ dialogue_dataset = load_dataset(
59
+ "nu-dialogue/real-persona-chat",
60
+ name="dialogue",
61
+ trust_remote_code=True
62
+ )
63
+ interlocutor_dataset = load_dataset(
64
+ "nu-dialogue/real-persona-chat",
65
+ name="interlocutor",
66
+ trust_remote_code=True
67
+ )
68
+ ```
69
+
70
+
71
+ ## Dataset Structure
72
+
73
+ ### Data Instances
74
+
75
+ 本コーパスは,対話データと話者データからなります.
76
+
77
+ 対話データには,対話ID,話者ID,発話,および,話者ごとの評価スコアが含まれます.評価スコアは,1が低いことを,5が高いことを示します.
78
+
79
+ ```json
80
+ {
81
+ "dialogue_id": 1,
82
+ "interlocutors": ["AA", "AB"],
83
+ "utterances": {
84
+ "utterance_id": [0, 1, 2, 3, 4, 5, ...],
85
+ "interlocutor_id": ["AA", "AB", "AA", "AB", "AA", "AB", ...],
86
+ "text": ["よろしくお願いいたします。", "よろしくお願いします!", "今日は涼しいですね", "雨が降って、何か涼しくなりましたね。", "そうですね、明日も涼しいと聞きました", "そうなんですか!でも、ちょっと湿度が高い気がします。", ...],
87
+ "timestamp": [datetime.datetime(2022, 8, 6, 14, 51, 18, 360000), datetime.datetime(2022, 8, 6, 14, 51, 48, 482000), datetime.datetime(2022, 8, 6, 14, 51, 55, 538000), datetime.datetime(2022, 8, 6, 14, 52, 07, 388000), datetime.datetime(2022, 8, 6, 14, 52, 16, 400000), datetime.datetime(2022, 8, 6, 14, 52, 31, 076000), ...]
88
+ },
89
+ "evaluations": {
90
+ "interlocutor_id": ["AA", "AB"],
91
+ "informativeness": [5, 5],
92
+ "comprehension": [5, 5],
93
+ "familiarity": [5, 5],
94
+ "interest": [5, 5],
95
+ "proactiveness": [5, 5],
96
+ "satisfaction": [5, 5]
97
+ }
98
+ }
99
+ ```
100
+
101
+ 話者データには,話者IDをキーとして,話者ID,ペルソナ,性格特性,属性,テキストチャットの経験が含まれます.性格特性スコアが高いほど,その性格的な傾向が強いことを示します.
102
+
103
+ ```json
104
+ {
105
+ "interlocutor_id": "AH",
106
+ "persona": [
107
+ "私は学生である。",
108
+ "埼玉県出身である。",
109
+ "私は毎日朝食を食べない。",
110
+ "私は毎日ウォーキングをする。",
111
+ "私はよくコンビニに行く。",
112
+ "私はタイピングが早い。",
113
+ "自分は物覚えが悪い。",
114
+ "自分は将来の目標が明確に決まっている。",
115
+ "毎日楽しいことを見つけられる。",
116
+ "自分は好きなものにはとことんこだわる。"
117
+ ],
118
+ "personality": {
119
+ "BigFive_Openness": 5.25,
120
+ "BigFive_Conscientiousness": 3.1666667461395264,
121
+ "BigFive_Extraversion": 3.3333332538604736,
122
+ "BigFive_Agreeableness": 4.166666507720947,
123
+ "BigFive_Neuroticism": 4.416666507720947,
124
+ "KiSS18_BasicSkill": 4.0,
125
+ "KiSS18_AdvancedSkill": 4.333333492279053,
126
+ "KiSS18_EmotionalManagementSkill": 4.0,
127
+ "KiSS18_OffenceManagementSkill": 4.0,
128
+ "KiSS18_StressManagementSkill": 4.0,
129
+ "KiSS18_PlanningSkill": 4.666666507720947,
130
+ "IOS": 4,
131
+ "ATQ_Fear": 5.0,
132
+ "ATQ_Frustration": 3.5,
133
+ "ATQ_Sadness": 3.0,
134
+ "ATQ_Discomfort": 3.3333332538604736,
135
+ "ATQ_ActivationControl": 3.7142856121063232,
136
+ "ATQ_AttentionalControl": 3.799999952316284,
137
+ "ATQ_InhibitoryControl": 3.142857074737549,
138
+ "ATQ_Sociability": 4.0,
139
+ "ATQ_HighIntensityPleasure": 4.5714287757873535,
140
+ "ATQ_PositiveAffect": 3.4000000953674316,
141
+ "ATQ_NeutralPerceptualSensitivity": 4.199999809265137,
142
+ "ATQ_AffectivePerceptualSensitivity": 4.400000095367432,
143
+ "ATQ_AssociativeSensitivity": 4.800000190734863,
144
+ "SMS_Extraversion": 2.5999999046325684,
145
+ "SMS_OtherDirectedness": 3.5833332538604736,
146
+ "SMS_Acting": 3.75
147
+ },
148
+ "demographic_information": {
149
+ "gender": 0,
150
+ "age": 0,
151
+ "education": 4,
152
+ "employment_status": 2,
153
+ "region_of_residence": 10
154
+ },
155
+ "text_chat_experience": {
156
+ "age_of_first_chat": 0,
157
+ "frequency": 0,
158
+ "chatting_partners": [0, 1],
159
+ "typical_chat_content": "学校に関すること、事務連絡など"
160
+ }
161
+ }
162
+ ```
163
+
164
+
165
+ ### Data Fields
166
+
167
+ 対話データの構造は次の通りです.
168
+ - `dialogue_id` (int32): 対話ID.
169
+ - `interlocutors` (sequence): 話者IDのリスト.
170
+ - `utterances` (sequence): 発話のリスト.
171
+ - `utterance_id` (int32): 発話ID.対話内で固有.0始まりのインデックス.
172
+ - `interlocutor_id` (string): 話者ID.
173
+ - `text` (string): 発話テキスト.
174
+ - `timestamp` (timestamp[us]): 発話終了時のタイムスタンプ.不明な場合は `0001-01-01T00:00:00.000000`.
175
+ - `evaluations` (sequence): 話者ごとの評価スコアのリスト.
176
+ - `interlocutor_id` (string): 話者ID.
177
+ - `informativeness` (int32): 情報量の評価スコア (1~5).
178
+ - `comprehension` (int32): 理解度の評価スコア (1~5).
179
+ - `familiarity` (int32): 親しみやすさの評価スコア (1~5).
180
+ - `interest` (int32): 興味の評価スコア (1~5).
181
+ - `proactiveness` (int32): 積極性の評価スコア (1~5).
182
+ - `satisfaction` (int32): 満足度の評価スコア (1~5).
183
+
184
+ 話者データの構造は次の通りです.
185
+ - `interlocutor_id` (strint): 話者ID.
186
+ - `persona` (sequence): ペルソナ.10文からなる.
187
+ - `personality` (struct): 性格特性.
188
+ - `BigFive_*` (float32): Big Five のスコア (1~7).
189
+ - `KiSS18_*` (float32): Kikuchi's Scale of Social Skills のスコア (1~5).
190
+ - `IOS` (int32): Inclusion of Others in the Self のスコア (1~7).
191
+ - `ATQ_*` (float32): Adult Temperament Questionnaire のスコア (1~7).
192
+ - `SMS_*` (float32): Self-Monitoring Scale のスコア (1~5).
193
+ - `demographic_information` (struct): 属性.
194
+ - `gender` (string): 性別.`Male`, `Female`, `Other` のいずれか.
195
+ - `age` (string): 年齢.`-19`, `20-29`, `30-39`, `40-49`, `50-59`, `60-69` のいずれか.
196
+ - `education` (string): 教育歴.`High school graduate`, `Two-year college`, `Four-year college`, `Postgraduate`, `Other` のいずれか.
197
+ - `employment_status` (string): 就労状況.`Employed`, `Homemaker`, `Student`, `Retired`, `Unable to work`, `None` のいずれか.
198
+ - `region_of_residence` (string): 居住地域.日本の都道府県名.
199
+ - `text_chat_experience` (struct): テキストチャットの経験.
200
+ - `age_of_first_chat` (string): 初めてテキストチャットをした時の年齢.`-9`, `10-19`, `20-29`, `30-39`, `40-49`, `50-59` のいずれか.
201
+ - `frequency` (string): 普段のテキストチャットの頻度.`Every day`, `Once every few days`, `Once a week`, `Less frequent than these` のいずれか.
202
+ - `chatting_partners` (sequence): 普段のテキストチャットの相手.`Family`, `Friend`, `Colleague`, `Other` のいずれか.
203
+ - `typical_chat_content` (string): 普段のテキストチャットの内容.
204
+
205
+ ### Data Splits
206
+
207
+ 対話データの分割は次の通りです.
208
+
209
+ | | train | validation | test |
210
+ | --- | --- | --- | --- |
211
+ | 対話数 | 13,583 | - | - |
212
+ | 発話数 | 408,619 | - | - |
213
+
214
+ 話者データの分割は次の通りです.
215
+
216
+ | | train | validation | test |
217
+ | --- | --- | --- | --- |
218
+ | 話者数 | 233 | - | - |
219
+
220
+
221
+ ### Citation Information
222
+
223
+ ```bibtex
224
+ @inproceedings{yamashita-etal-2023-realpersonachat,
225
+ title = "{R}eal{P}ersona{C}hat: A Realistic Persona Chat Corpus with Interlocutors{'} Own Personalities",
226
+ author = "Yamashita, Sanae and
227
+ Inoue, Koji and
228
+ Guo, Ao and
229
+ Mochizuki, Shota and
230
+ Kawahara, Tatsuya and
231
+ Higashinaka, Ryuichiro",
232
+ booktitle = "Proceedings of the 37th Pacific Asia Conference on Language, Information and Computation",
233
+ year = "2023",
234
+ pages = "852--861"
235
+ }
236
+
237
+ @inproceedings{yamashita-etal-2024-realpersonachat-ja,
238
+ title = "{R}eal{P}ersona{C}hat: 話者本人のペルソナと性格特性を含んだ雑談対話コーパス",
239
+ author = "山下 紗苗 and 井上 昂治 and 郭 傲 and 望月 翔太 and 河原 達也 and 東中 竜一郎",
240
+ booktitle = "言語処理学会第30回年次大会発表論文集",
241
+ year = "2024",
242
+ pages = "2738--2743"
243
+ }
244
+ ```
245
+
246
+
247
+ ## Acknowledgment
248
+
249
+ 本コーパスは,[JSTムーンショット型研究開発事業,JPMJMS2011](https://www.avatar-ss.org/) の支援を受けて構築しました.
250
+
251
+ <img src="https://github.com/nu-dialogue/real-persona-chat/blob/main/Moonshot_Logo.png?raw=true" alt="ムーンショットのロゴ" style="max-width: 100%; width: 300px;">
252
+
253
+
254
+ ## License
255
+
256
+ 本コーパスは [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/deed.ja) の下で提供されます.
257
+
258
+ [![CC BY-SA 4.0](https://i.creativecommons.org/l/by-sa/4.0/88x31.png)](https://creativecommons.org/licenses/by-sa/4.0/deed.ja)