Dataset Preview
Viewer
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
An error occurred while generating the dataset All the data files must have the same columns, but at some point there are 2 missing columns ({'intentClass', 'slotClasses'}) This happened while the csv dataset builder was generating data using hf://datasets/Matrix430/CONDA/test.csv (at revision 31f2cdc6bff5ad6def58e07d3c3549dc393bc086) Please either edit the data files to have matching columns, or separate them into different configurations (see docs at https://hf.co/docs/hub/datasets-manual-configuration#multiple-configurations)
Error code:   UnexpectedError

Need help to make the dataset viewer work? Open a discussion for direct support.

Id
int64
utterance
string
tokenized
string
intentClass
string
slotClasses
string
matchId
int64
conversationId
int64
chatTime
int64
11,263
wow!
wow
O
O
697
3,193
76
13,741
WTF
WTF
O
T
843
3,809
1,563
22,125
wpe wpe
wpe wpe
O
O O
1,412
6,199
2,853
6,453
hahaha
hahaha
O
O
439
1,875
1,038
9,644
wtf
wtf
O
T
601
2,713
1,661
6,208
i cant [SEPA] play [SEPA] with 4 trash
i cant [SEPA] play [SEPA] with 4 trash
E
P O SEPA O SEPA O O O
433
1,836
1,674
38,968
bg
bg
O
O
2,572
10,979
2,875
3,213
=) [SEPA] ty
#ERROR!
O
O
240
999
1,797
44,326
gg [SEPA] report my team rat [SEPA] please
gg [SEPA] report my team rat [SEPA] please
A
S SEPA S P O S SEPA O
3,001
12,737
2,602
30,888
ez mid
ez mid
I
S S
2,020
8,676
3,044
4,993
hahah
hahah
O
O
343
1,469
2,000
5,736
my arrows [SEPA] always decent [SEPA] fuck u
my arrows [SEPA] always decent [SEPA] fuck u
E
P O SEPA O O SEPA T P
408
1,717
1,955
33,772
gh
gh
O
O
2,229
9,529
1,071
37,524
always engage at bot, lc cant takle bot then top
always engage at bot lc cant takle bot then top
O
O O O S D O O S O S
2,464
10,554
1,165
36,088
cmon
cmon
O
O
2,366
10,203
2,833
33,901
the comeback is real. :)
the comeback is real :)
O
O O O O O
2,236
9,556
2,207
21,047
him mid vs me
him mid vs me
O
P S O P
1,336
5,855
2,572
43,817
just end [SEPA] i wan nex game
just end [SEPA] i wan nex game
O
O O SEPA P O O O
2,961
12,556
2,617
36,694
g?
g
A
O
2,411
10,330
-68
44,212
he is not losing
he is not losing
O
P O O O
2,994
12,700
2,001
43,049
Pls report sb thanks
Pls report sb thanks
A
O S C O
2,900
12,271
864
39,334
omg
omg
O
O
2,617
11,133
2,605
10,429
ggwp
ggwp
O
S
654
2,960
3,772
1,874
WP cap
WP cap
O
S O
149
613
2,208
14,416
lo [SEPA] lol
lo [SEPA] lol
O
O SEPA O
878
3,972
-6
8,653
fuyckjerfe
fuyckjerfe
E
O
532
2,399
1,057
34,988
gg noob mid invoker
gg noob mid invoker
E
S T S C
2,314
9,898
989
7,644
i mean not everyone on ur team is dumb enough to rot down to 20% hp
i mean not everyone on ur team is dumb enough to rot down to 2 hp
E
P O O P O P O O T P O D O O O O
491
2,154
706
32,480
8 7?
8
O
O
2,126
9,121
2,755
33,025
it just random
it just random
O
P O O
2,173
9,307
1,434
650
:/
:/
O
O
59
222
1,726
30,513
u dead [SEPA] .i.
u dead [SEPA] i
O
P T SEPA P
1,986
8,555
1,588
35,218
he just keeps charhing [SEPA] since spectre comited ult [SEPA] so did I :D [SEPA] i thought they will go for more blood [SEPA] but whatever
he just keeps charhing [SEPA] since spectre comited ult [SEPA] so did I :D [SEPA] i thought they will go for more blood [SEPA] but whatever
O
P O O O SEPA O C O S SEPA O O P O SEPA P O P O S O O S SEPA O P
2,324
9,954
2,096
7,740
me too
me too
O
P O
493
2,168
2,165
27,300
gg ty for mmr
gg ty for mmr
O
S O O D
1,760
7,635
1,758
34,870
gg
gg
O
S
2,307
9,851
1,898
34,322
you know you can take the mana thing out of the mana boots
you know you can take the mana thing out of the mana boots
O
P O P O S O S O O O O S D
2,270
9,692
1,984
24,708
u gave me free farm [SEPA] u could kill me 100 times
u gave me free farm [SEPA] u could kill me 100 times
O
P O P O S SEPA P O O P O O
1,579
6,939
1,562
21,208
wp [SEPA] legoin report
wp [SEPA] legoin report
A
S SEPA O S
1,348
5,900
5,448
37,184
WP m8 [SEPA] nice game but too many shit talk [SEPA] thx sea [SEPA] too many cancer talktalktalk
WP m8 [SEPA] nice game but too many shit talk [SEPA] thx sea [SEPA] too many cancer talktalktalk
O
S O SEPA O O O O P T O SEPA O O SEPA O P S O
2,443
10,453
2,740
14,480
SHIT
SHIT
E
T
881
3,989
1,776
7,000
talking to my boy alch
talking to my boy alch
O
O O P O C
462
2,005
1,764
13,589
TIME FOR A BIT OF RO SHAM BO
TIME FOR A BIT OF RO SHAM BO
O
O O O O O O O O
833
3,767
1,356
42,883
COMEND DOOM IN LEGION
COMEND DOOM IN LEGION
A
O C O C
2,888
12,219
1,630
17,055
greedy
greedy
O
O
1,050
4,719
609
22,468
dont def [SEPA] noob team
dont def [SEPA] noob team
E
O O SEPA T O
1,428
6,291
2,134
42,423
gg
gg
O
S
2,855
12,072
1,568
24,740
spectre isnt meant to win the lane
spectre isnt meant to win the lane
O
C O O O O O S
1,580
6,947
3,685
257
shit [SEPA] SHIT [SEPA] SHIT [SEPA] SHITTT! :)
shit [SEPA] SHIT [SEPA] SHIT [SEPA] SHITTT :)
E
T SEPA T SEPA T SEPA O O
29
105
2,926
18,966
sad bs
sad bs
O
O S
1,178
5,283
1,916
3,045
ez
ez
I
S
227
956
2,542
11,994
T T QOP
T T QOP
O
O O C
749
3,427
-59
20,971
you guys win this fag went mid [SEPA] so fucking bad mid
you guys win this fag went mid [SEPA] so fucking bad mid
E
P P O P T O S SEPA O T O S
1,331
5,833
2,132
34,245
Alche and lycan
Alche and lycan
O
C O C
2,262
9,652
2,062
33,681
Rip
Rip
O
O
2,226
9,524
827
34,563
singing siren ,dancing jugg
singing siren dancing jugg
O
O O O C
2,284
9,756
1,645
27,700
ez game
ez game
I
S O
1,786
7,748
2,769
26,479
meeporino feederino
meeporino feederino
O
C O
1,705
7,419
1,246
1,939
ur asshole
ur asshole
E
P T
153
622
1,990
4,236
Thats all I gotta say [SEPA] Drow 2-5
Thats all I gotta say [SEPA] Drow 5
I
P P P O O SEPA C O
310
1,315
1,534
40,002
COMMEND
COMMEND
A
O
2,651
11,331
2,440
35,761
lowskill shits [SEPA] all you have to do is dive endlessly [SEPA] and this cunt silencer and invoekr [SEPA] let you win
lowskill shits [SEPA] all you have to do is dive endlessly [SEPA] and this cunt silencer and invoekr [SEPA] let you win
E
O T SEPA P P O O O O S O SEPA O P T C O C SEPA O P O
2,352
10,113
1,975
40,923
tnx for the russian team obama....could have at least nuked putin
tnx for the russian team obama could have at least nuked putin
E
O O O O O O O O O O O O
2,727
11,630
-27
2,468
wise ffs [SEPA] can youi stop inviting strangers to fuck your grandmothers' corpse ?
wise ffs [SEPA] can youi stop inviting strangers to fuck your grandmothers corpse
E
O T SEPA O O O O O O T P O O
183
761
-40
23,502
nice
nice
O
O
1,498
6,621
1,691
34,128
heng i never die
heng i never die
E
O P O O
2,257
9,621
2,237
6,531
We all randomed
We all randomed
O
P P O
441
1,891
-63
36,835
golden medal for u lc
golden medal for u lc
O
O O O P D
2,413
10,362
2,787
27,604
GG lag
GG lag
O
S O
1,780
7,715
2,105
4,223
HAHA
HAHA
O
O
310
1,312
840
37,287
btw AM nice blink into stun :D
btw AM nice blink into stun :D
O
O C O D O S O
2,453
10,485
758
39,369
who care [SEPA] ?
who care [SEPA]
O
P S SEPA
2,621
11,145
615
14,315
emergency plan sry
emergency plan sry
O
O O O
875
3,946
37
17,067
i want slardar getting strong
i want slardar getting strong
O
P O C O O
1,050
4,722
1,014
41,937
ahahaha
ahahaha
O
O
2,801
11,903
954
41,525
ez tho
ez tho
I
S O
2,764
11,769
2,075
7,472
i know your feeling bro
i know your feeling bro
O
P O P O P
484
2,121
2,548
23,927
whos your daddy
whos your daddy
O
O P O
1,526
6,742
1,316
15,476
stunning
stunning
O
O
951
4,261
3,010
37,103
5 man doto
5 man doto
O
O P O
2,436
10,436
2,080
34,107
gg wp [SEPA] Sry for my feed
gg wp [SEPA] Sry for my feed
O
S S SEPA O O P D
2,256
9,616
2,056
5,094
dnt play storm
dnt play storm
I
O O C
348
1,488
2,400
13,915
^LOL
LOL
O
O
852
3,863
2,130
15,986
this couldn't be won
this could not be won
O
P O O O O
988
4,399
2,015
42,866
gg wp
gg wp
O
S S
2,887
12,210
3,914
14,307
weaver mid dont worry
weaver mid dont worry
O
C S O O
875
3,946
-36
25,624
trone
trone
O
S
1,643
7,190
1,978
10,631
./. [SEPA] bm?
[SEPA] bm
O
SEPA C
662
2,994
1,238
3,620
gj guys...
gj guys
O
S P
271
1,107
3,139
12,706
gg wp
gg wp
O
S S
781
3,569
4,119
36,155
Ty
Ty
O
O
2,373
10,227
2,166
43,310
so ez [SEPA] ahahaha
so ez [SEPA] ahahaha
I
O S SEPA O
2,919
12,371
1,599
28,846
hahahai
hahahai
O
O
1,859
8,065
453
27,595
and blame us
and blame us
O
O O P
1,780
7,710
1,123
37,630
YEAAHHHHHHHHH [SEPA] GGGGGGGGGGGG
YEAAHHHHHHHHH [SEPA] GGGGGGGGGGGG
O
O SEPA S
2,470
10,588
3,173
39,425
EZ
EZ
I
S
2,624
11,160
716
26,080
aw
aw
O
O
1,669
7,269
136
32,889
and telling im too much of a solo player
and telling im too much of a solo player
O
O O O O O O O S O
2,161
9,260
2,383
43,288
He's never picked in pubs, and the only bracket he has over 50% win rate is 'very high skill' on asian servers. [SEPA] So shut up and pick something fun. [SEPA] You cucks.
He s never picked in pubs and the only bracket he has over 5 win rate is very high skill on asian servers [SEPA] So shut up and pick something fun [SEPA] You cucks
E
P O O O O O O O O O P O O O O O O O O S O O O SEPA O O O O O P O SEPA P O
2,919
12,369
1,015
38,667
2v5 game
2v5 game
O
O O
2,540
10,886
2,572
End of preview.

Dataset Card for CONDA

Dataset Summary

Traditional toxicity detection models have focused on the single utterance level without deeper understanding of context. We introduce CONDA, a new dataset for in-game toxic language detection enabling joint intent classification and slot filling analysis, which is the core task of Natural Language Understanding (NLU). The dataset consists of 45K utterances from 12K conversations from the chat logs of 1.9K completed Dota 2 matches. We propose a robust dual semantic-level toxicity framework, which handles utterance and token-level patterns, and rich contextual chatting history. Accompanying the dataset is a thorough in-game toxicity analysis, which provides comprehensive understanding of context at utterance, token, and dual levels. Inspired by NLU, we also apply its metrics to the toxicity detection tasks for assessing toxicity and game-specific aspects. We evaluate strong NLU models on CONDA, providing fine-grained results for different intent classes and slot classes. Furthermore, we examine the coverage of toxicity nature in our dataset by comparing it with other toxicity datasets.

Leaderboards

The Codalab leaderboard can be found at: https://codalab.lisn.upsaclay.fr/competitions/7827

Evaluation Metrics

JSA(Joint Semantic Accuracy) is used for ranking. An utterance is deemed correctly analysed only if both utterance-level and all the token-level labels including Os are correctly predicted.

Besides, the f1 score of utterance-level E(xplicit) and I(mplicit) classes, token-level T(oxicity), D(ota-specific), S(game Slang) classes will be shown on the leaderboard (but not used as the ranking metric).

Languages

English

Video

Please enjoy a video presentation covering the main points from our paper:

ACL_video

Citation Information

@inproceedings{weld-etal-2021-conda,
    title = "{CONDA}: a {CON}textual Dual-Annotated dataset for in-game toxicity understanding and detection",
    author = "Weld, Henry  and
      Huang, Guanghao  and
      Lee, Jean  and
      Zhang, Tongshu  and
      Wang, Kunze  and
      Guo, Xinghong  and
      Long, Siqu  and
      Poon, Josiah  and
      Han, Caren",
    booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
    month = aug,
    year = "2021",
    address = "Online",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2021.findings-acl.213",
    doi = "10.18653/v1/2021.findings-acl.213",
    pages = "2406--2416",
}
Downloads last month
2
Edit dataset card