MADOC: Bluesky / bluesky

Every post, comment and repost collected from Bluesky between 28 June 2023 and 18 March 2024, as published in the MADOC dataset (Multi-Platform Aggregated Dataset of Online Communities). The package holds 2,814,352 interactions (971,431 reposts, 932,545 comments, 910,376 posts) by 217,768 pseudonymised users. Each interaction has its text, the ID of the post or comment it replies to, and four machine-computed scores: VADER sentiment, TextBlob sentiment, TextBlob subjectivity and a ToxiGen toxicity score. MADOC publishes one file per community and platform; this package is the bluesky_madoc.parquet file.

#Icwsm#Madoc#Reply network#Toxigen#Vader

Download data package README Original files as received BibTeX JSONSimilar datasets

Analytics

The interactive network needs JavaScript.

· graph.json

The full graph has 164,265 nodes; this view shows the 1,000-node sample the pipeline drew for display. Node size is centrality, colour is community.

164.3K
Nodes
467.7K
Edges
1.73e-05
Density
3.9K
Components
-0.02394
Assortativity
1.9%
Reciprocity
86K
Bridges
Most central nodes
e7263999-68b6-5a23-b530-af25b7efd6320.05845
f4e5e9c8-3888-52ca-990c-a588d14f69d80.04061
c4b445d5-79ce-5e9f-a51b-579b53e859ac0.04
a345b4cc-b02a-5aa7-88b1-2f50ac5aecc90.03622
3e0bc386-a8c7-58cd-8065-bf5faa5b30550.03132
1ca4a789-9f15-5e59-bcdb-e51bd5cdbe9c0.02758
a5c4208f-1d01-59ea-bf1c-65786d07d37d0.02746
ffcd73aa-1a89-5b3e-b3a0-a146747ea9530.02702
28aefb74-f055-586c-98ce-6970531ddbeb0.02596
df2ad546-a4f0-51ac-b38c-88216742e5530.02439
Content metrics

2,814,352 records scored · min–maxmean ± 0.675 σ (the interquartile range if the values are roughly normal)mean

x strict filter21.2% true
true 21.2%false 78.8%
x sentiment vaderμ 0.088 ± 0.496
min -1max 1
low 30%medium 37%high 33%
x sentiment textblobμ 0.092 ± 0.289
min -1max 1
low 23%medium 44%high 33%
x subjectivity textblob40.9% true
true 40.9%false 59.1%
x toxicity toxigenμ 0.122 ± 0.281
min 0.001max 0.993
low 33%medium 34%high 33%
8.4
Flesch–Kincaid grade
US school grade level
9.4
Automated readability index
Frequent terms
likepeopledontimegoodknownewtwitterthinkwantgotartgoingshitdayloverightneedworkwaythingtrumptodayyearguypostyearslolthingsnsfwlotlittleletfeelgreatactuallyuselookbadgettingbigcovidcathateblueskylongbetterfuckingoldsaid
Topic clusters

Topic 0

nsfw
furryart
art
furry
illustration
draw
nsfwart
patreon

Topic 1

covid
vaccine
mask
cdc
test
flu
virus
pandemic

Topic 2

art
repost
raffle
winner
draw
artist
character
follow

Topic 3

look
laugh
find
clab
day
come
tell
wait

Topic 4

trump
election
biden
republicans
voter
desantis
democrats
donald

Topic 5

great
amazing
good
lovely
congratulation
incredible
excellent
luck

Topic 6

israel
gaza
hamas
israeli
palestinian
palestinians
palestine
war

Topic 7

til
som
det
bara
inn
den
hér
han

Topic 8

twitter
post
account
user
follower
site
engagement
platform

Topic 9

cat
kitty
loaf
big
dog
feed
lackadaisy
sit

Topic 10

people
black
like
racism
argument
think
mean
thing

Topic 11

twitter
tweet
mutual
tumblr
delete
follow
account
text
Show all 70 topics

Topic 12

feel
bark
like
walk
eat
house
love
sleep

Topic 13

trump
court
donald
vote
supreme
voter
election
president

Topic 14

lol
happen
think
nope
lmao
loop
surprised
mean

Topic 15

trans
cis
tran
people
queer
transphobic
transphobia
ally

Topic 16

bluesky
twitter
bsky
welcome
gif
account
invite
sky

Topic 17

problem
solution
jerk
issue
talk
drive
dogwhistle
hear

Topic 18

cool
puppy
dog
peak
thicc
look
catch
smooch

Topic 19

russia
ukraine
putin
russian
nato
ukrainian
navalny
drone

Topic 20

fuck
jesus
christ
fucking
nobel
fuckin
prize
ffs

Topic 21

cheese
pizza
breakfast
eat
butter
soup
grill
mac

Topic 22

musk
elon
worker
union
breaking
tesla
twitter
file

Topic 23

meme
repost
reskeet
pic
favourite
reskeete
goose
film

Topic 24

cute
cutie
adorable
awww
thats
pie
fluffy
hello

Topic 25

movie
pleasant
film
wake
time
minute
play
start

Topic 26

marvel
finish
movie
dungeon
youtube
release
podcast
pokemon

Topic 27

hear
absolutely
agree
feel
definitely
sure
lol
fully

Topic 28

game
play
mario
controller
console
nintendo
quake
video

Topic 29

bluesky
feed
post
guide
label
hashtag
tab
repost

Topic 30

shit
holy
trouble
white
bombshell
chap
gokai
caveman

Topic 31

know
people
try
apology
bad
talk
shit
kid

Topic 32

year
fucking
pay
shit
tedious
tablet
like
purchase

Topic 33

hey
ooh
owo
goooooo
ohhh
goooo
oooh
woah

Topic 34

like
lick
ball
nymph
cloth
dangerously
lightbulb
disturb

Topic 35

help
donate
aid
paypal
mutual
need
donation
goal

Topic 36

hope
soon
feel
tomorrow
mate
good
derail
argh

Topic 37

sorry
apologize
loss
horrible
fair
fuck
hear
dreadful

Topic 38

tonight
watch
rugby
movie
song
christmas
winter
day

Topic 39

know
boston
app
apply
paint
lol
lectern
bleach

Topic 40

sauce
dinner
cheese
salad
chicken
bread
toast
garlic

Topic 41

dog
hawk
bird
pup
golden
hitman
red
retriever

Topic 42

meow
mow
dun
maw
hop
babe
hot
phat

Topic 43

remember
forget
funny
memory
lol
forgot
post
place

Topic 44

suck
wrong
incomprehensible
incapable
simply
inconsistent
unfuckable
inconceivable

Topic 45

hate
false
hater
dislike
enemy
dictionary
brighton
rebuttal

Topic 46

racist
sunak
hate
meaningless
bankruptcy
depressing
permanent
leftist

Topic 47

cat
kitten
kitty
korra
foster
window
eye
leg

Topic 48

yes
yeah
guilty
shack
aight
afab
bonker
space

Topic 49

big
guy
little
dick
gang
huge
girl
strong

Topic 50

thank
chance
opportunity
justin
goodness
god
share
inventive

Topic 51

love
omg
style
outer
pdx
jester
cackle
chandler

Topic 52

dream
favorite
rotation
blunt
dreams
nightmare
best
prefer

Topic 53

trans
gender
cis
people
writer
transgender
identity
neato

Topic 54

tax
gun
housing
stanford
pay
talk
adjacent
deprive

Topic 55

right
correct
real
sike
emerald
lectern
spatial
mcdonalds

Topic 56

app
web
user
composer
plugin
verification
bluesky
alert

Topic 57

coach
cringe
trolley
driving
bond
sir
description
stop

Topic 58

jews
israel
jewish
israeli
palestine
nancy
synagogue
netanyahu

Topic 59

clean
pepsi
want
damn
need
wanna
wtf
chips

Topic 60

guy
blond
far
insane
forever
gaius
ritchie
jquery

Topic 61

truth
true
far
minnesota
think
story
away
flag

Topic 62

evil
horrible
resident
bad
terrible
spy
tis
carlos

Topic 63

sleep
bed
morning
uneasy
caffeine
transform
wake
awake

Topic 64

fascist
fascism
nazi
sunak
rishi
rwanda
summit
host

Topic 65

nazi
nazis
idea
marketplace
debate
defeat
bad
people

Topic 66

morning
hope
good
weekend
birthday
happy
day
enjoy

Topic 67

commission
twitter
view
comm
portrait
telegram
patreon
loogamon

Topic 68

birthday
happy
cake
quill
slipper
zap
android
comfy

Topic 69

beautiful
stunning
busty
bpd
scenery
beyoncé
jin
ahh
Top hashtags
#nsfw37.6K
#furryart34.2K
#art27.2K
#furry23.2K
#illustration4.7K
#pokemon4.6K
#vtuber3.5K
#digitalart3.1K
#nsfwart2.4K
#furrynsfw2.3K
#oc2.3K
#fanart2.2K
#writingcommunity2K
#fursuit2K
#adsb2K
#pnw2K
#ksea2K
#piaware2K
#seatac2K
#11.9K
Top emoji
😂14K
😭12.4K
💜11.6K
👀9.9K
💙9.9K
🎨9.3K
🤣9K
🥰8.9K
💕7K
🔥6.9K
💦6.8K
💖6.4K
😍5.9K
🙏5.7K
👇5.7K

Computed over the full corpus.

Record

Catalog key
madoc-bluesky-bluesky
Authors
Dankulov, Marija Mitrović; Tomašević, Aleksandar; Maletić, Slobodan; Anđelković, Miroslav; Vranić, Ana; Cvetković, Darja; Stupovski, Boris; Vudragović, Dušan; Major, Sara; Bogojević, Aleksandar
Year
2025
Source paper
MADOC: Multi-Platform Aggregated Dataset of Online Communities
DOI / arXiv
10.1609/icwsm.v19i1.35954
License
Creative Commons Attribution 4.0 International (CC BY 4.0) (CC-BY-4.0)
Records
2,795,981
Nodes
164,265
Edges
467,653 (directed)
Nodes are
user
Edges are
reply
Attributes
nodes have no attributes; edges have no attributes
Text corpus
post text
Languages
en, es, ja, pt
Collected
2012 to 2024 (2010s, 2020s)
Tasks
From the graph: link prediction, community detection
Region
Global (inferred: the platform is used worldwide)

Provenance

Originally published at
https://zenodo.org/records/15690964 link checked
Original files as received
The source files, unconverted, kept for reference: mirrored on echo.directory on 2026-10-02, 455.3 MB across 1 files. Checksums in MANIFEST.json. For use, start with the data package below.
Data package
Version 1.0.0 of this dataset as an ECHO Data Package: posts (2,814,352 rows).
Structure
Temporal (timestamped), Text, Threaded conversations

How to cite

BibTeX

@misc{echo2026,
  title  = {ECHO: Exploratory Compendium of Humanity Online},
  author = {Switzer, Braden and Zafarani, Reza},
  year   = {2026},
  url    = {https://echo.directory},
  note   = {Syracuse University}
}

@article{dankulov2025madoc,
  title   = {{MADOC: Multi-Platform Aggregated Dataset of Online Communities}},
  author  = {Dankulov, Marija Mitrović and Tomašević, Aleksandar and Maletić, Slobodan and Anđelković, Miroslav and Vranić, Ana and Cvetković, Darja and Stupovski, Boris and Vudragović, Dušan and Major, Sara and Bogojević, Aleksandar},
  journal = {Proceedings of the International AAAI Conference on Web and Social Media (ICWSM)},
  volume  = {19},
  year    = {2025},
  doi     = {10.1609/icwsm.v19i1.35954}
}

APA

Switzer, B., & Zafarani, R. (2026). ECHO: Exploratory Compendium of Humanity Online [Data repository]. Syracuse University. https://echo.directory

Dankulov, M. M., Tomašević, A., Maletić, S., Anđelković, M., Vranić, A., Cvetković, D., Stupovski, B., Vudragović, D., Major, S., & Bogojević, A. (2025). MADOC: Bluesky / bluesky [Data set]. https://doi.org/10.1609/icwsm.v19i1.35954

MLA

Switzer, Braden, and Reza Zafarani. ECHO: Exploratory Compendium of Humanity Online. Syracuse University, 2026, echo.directory.

Dankulov, Marija Mitrović, et al. MADOC: Bluesky / bluesky. 2025, doi.org/10.1609/icwsm.v19i1.35954.

BibTeX · RIS · CSL-JSON

Similar datasets 10 all with scores →