word1
stringlengths 2
29
| word2
stringlengths 2
31
| similarity
float32 0
4
|
|---|---|---|
Pokemon
|
Pocket_Monsters
| 3.81
|
prejudice
|
chauvinist
| 2.25
|
formic_acid
|
arachnology
| 1.19
|
NetMeeting
|
Marwar_Hall
| 0
|
kingfish
|
kingship
| 0.31
|
iight
|
ok
| 3.94
|
ACL
|
EMNLP
| 3.13
|
Qintex
|
Allwaste
| 2.06
|
Australian_Open
|
mixed_doubles
| 1.88
|
Curry_powder
|
pumpkin_spice
| 2.75
|
full-HD
|
1080p
| 4
|
cheddah
|
cheddar
| 0.25
|
convocation
|
gathering
| 3.56
|
random_seed
|
BiLSTM
| 1.56
|
heater
|
convector
| 3.5
|
half-life
|
ratemeter
| 1.81
|
3D
|
black-and-white
| 1.13
|
Hero's_engine
|
aeolipile
| 4
|
Josef_Albers
|
Richard_Anuszkiewicz
| 2.06
|
MacBook
|
ZenBook
| 3.13
|
microwaving
|
pesto
| 0.75
|
primality
|
mathematics
| 2
|
Park_Ji-sung
|
Yosemite_Park
| 0
|
router
|
D-Link
| 2.13
|
Winamp
|
VLC_media_player
| 3.19
|
gown
|
pelerine
| 2.38
|
Malva_parviflora
|
cheeseweed
| 4
|
care
|
caution
| 3
|
rope-a-dope
|
WWE
| 1.81
|
oldster
|
dotard
| 3.31
|
navaid
|
HIV/Aids
| 0
|
excellent
|
top-notch
| 3.81
|
MIDlet
|
Oracle_Java
| 2.25
|
fakelore
|
photostimulation
| 0.06
|
black_hole
|
blackmail
| 0.06
|
night_sky
|
skyglow
| 1.94
|
neuropore
|
nervous_system
| 2.31
|
underspecification
|
incompleteness
| 2.94
|
exequatur
|
equator
| 0
|
Rotary_International
|
Rota_Island
| 0.06
|
TorPark
|
parkour
| 0
|
yellow_dwarf
|
yellow_pages
| 0.06
|
Mercedes-Benz
|
BMW
| 3.13
|
circus
|
ropedancer
| 2
|
Hoover_hog
|
armadillo
| 3.94
|
irresistibleness
|
illiterateness
| 0.13
|
metolazone
|
blazonry
| 0
|
vasocongestion
|
engorgement
| 2.88
|
salicylic_acid
|
carbonate
| 2.31
|
Kepler-11
|
red_giant
| 2.56
|
baby
|
cutee
| 1.56
|
Rubik's_Cube
|
RuBisCO
| 0.06
|
goatsbeard
|
tragopogon
| 4
|
fundraiser
|
event
| 2
|
going
|
really
| 0
|
transmigration
|
residence_permit
| 2.06
|
prospector
|
sourdough
| 3.56
|
sorry
|
srry
| 4
|
avionics
|
aeronautics
| 2.5
|
rallentando
|
slowly
| 2.88
|
retweeting
|
RTing
| 4
|
Apple
|
Applebees
| 0.38
|
exponential
|
logx
| 2.38
|
Zeta-Jones
|
Catherine_Zeta-Jones
| 3.88
|
Mosul
|
Mawsil
| 4
|
Pizza_Hut
|
Pizzle_rot
| 0.06
|
remainder
|
difference
| 2.44
|
preheat
|
reheat
| 2.63
|
disembodied
|
spiritual
| 2.56
|
crested_tit
|
Amazon_rainforest
| 0.94
|
afterworld
|
purgatory
| 2.88
|
screenshot
|
screengrab
| 3.88
|
practicable
|
goal
| 1.13
|
Skype_Lite
|
ooVoo
| 2.75
|
decomposition
|
factorization
| 3.31
|
LOL
|
looool
| 3.88
|
skateboard_deck
|
halfpipe
| 2.63
|
Breuil-Cervinia
|
Val_Gardena
| 3.06
|
inheritor
|
hoarded_wealth
| 1.94
|
appendage
|
swimmeret
| 2.69
|
passenger
|
passepied
| 0.06
|
tedious
|
old-fashioned
| 1
|
radionavigation
|
frequency_band
| 1.75
|
Tag_Heuer
|
Jaeger-LeCoultre
| 3.13
|
Followback
|
Twitter
| 2.31
|
weekend
|
race
| 0.13
|
septenary
|
Pleiades
| 1.25
|
monsignor
|
assignor
| 0.25
|
preoccupation
|
prepossession
| 1.19
|
spontaneousness
|
returnability
| 0.31
|
under-appreciated
|
unnoticeable
| 1.69
|
covfefe
|
coverage
| 2.69
|
devious
|
untrustworthy
| 2.75
|
comedian
|
stand-up
| 2.44
|
infant
|
breastfeeder
| 2.19
|
trusteeship
|
traineeship
| 0.31
|
human_face
|
make-up
| 1.88
|
backslash
|
backsolving
| 0.19
|
cr8
|
create
| 3.94
|
fluoride
|
monofluoride
| 3.31
|
End of preview. Expand
in Data Studio
Card-660 Dataset
This is the Card-660 (Cambridge Rare Word) dataset, converted to Hugging Face datasets format.
Description
Card-660 is a benchmark for evaluating semantic similarity models, specifically focusing on rare and infrequent words. It contains 660 word pairs annotated by experts. It was designed to solve the issues of low inter-annotator agreement found in other rare-word datasets (like RW).
Columns
word1: First word of the pairword2: Second word of the pairsimilarity: Human-annotated similarity score (scale typically 0-4 or 0-10)
Usage
from datasets import load_dataset
dataset = load_dataset("Yuti/Card-660")
- Downloads last month
- 30