Hugging Face's logo

Datasets:
almogtavor
/
stanford-rare-word-similarity-dataset

Modalities:
Tabular
Text
Formats:
csv
Size:
1K - 10K
Tags:
fasttext
Libraries:
Datasets
pandas
License:
Dataset card Data Studio Files
xet
 Community
Dataset Viewer
Auto-converted to Parquet Duplicate
Word1
stringlengths
7
18
Word2
stringlengths
2
20
Avg_Similarity
float64
0
10
Rating1
int64
0
10
Rating2
int64
0
10
Rating3
int64
0
10
Rating4
int64
0
10
Rating5
int64
0
10
Rating6
int64
0
10
Rating7
int64
0
10
Rating8
float64
0
10
⌀
Rating9
float64
0
10
⌀
Rating10
float64
0
10
⌀
squishing
squirt
5.88
7
7
6
1
4
6
6
7
2
4
undated
undatable
5.83
10
9
6
5
5
7
7
9
2
5
circumvents
beat
5.33
7
7
3
9
8
6
3
2
0
6
circumvents
ebb
3.25
7
4
6
4
2
0
0
3
6
0
dispossess
deprive
6.83
10
6
7
10
10
8
5
8
1
7
provincialism
narrow-mindedness
8.11
9
7
9
9
8
6
9
7
0
9
provincialism
partiality
4.5
9
3
0
4
8
9
6
5
0
0
instrumentality
department
3
6
2
4
3
7
3
3
0
0
0
instrumentality
utility
7.29
6
7
7
5
8
7
9
8
3
8
involvement
action
6.86
7
8
7
8
5
9
6
9
7
1
involvement
implication
4.17
5
8
2
3
10
8
4
4
4
5
ecclesiastic
clergyman
8.67
10
8
10
10
8
6
8
0
9
9
brigadier
general
8.17
8
8
10
8
9
7
9
6
5
10
carbonic
chemical
5.17
7
3
5
4
8
7
9
0
2
5
carbonic
paper
5.17
8
7
6
2
8
7
7
0
0
2
aspirate
pronounce
5.2
8
4
2
10
7
6
0
0
7
0
aspirate
remove
2.5
6
3
7
3
2
0
2
0
null
null
monotype
machine
5.25
1
8
1
6
1
5
5
9
0
5
incommensurate
incommensurable
7.6
10
10
8
7
9
8
4
6
null
null
campfires
fire
9.33
10
9
9
9
9
10
10
8
10
6
cognizance
knowing
9.44
10
9
10
8
10
10
8
10
10
7
urbanize
change
5.67
4
2
2
7
5
7
4
10
2
7
imperfection
state
0.29
0
1
0
0
1
0
6
4
0
null
assessment
charge
4.4
0
8
2
1
7
8
5
7
0
null
assessment
assay
5.83
8
10
5
6
4
5
7
3
null
null
incubate
breed
6.4
10
6
10
5
7
4
7
7
null
null
incubate
develop
7.2
7
7
8
6
7
7
9
10
null
null
principality
domain
7.62
7
8
8
4
8
0
7
8
7
8
vicarious
abnormal
0.12
3
0
0
0
5
0
0
0
0
1
vicarious
secondary
6.5
7
8
0
9
5
0
10
8
2
null
ungraceful
awkward
8.5
10
7
9
9
10
7
8
10
7
8
unsighted
color-blind
5.29
6
5
7
4
6
5
5
0
6
null
socialise
educate
6.12
6
7
1
7
5
7
6
4
7
0
socialise
swing
1
5
6
1
3
0
4
0
0
0
0
diagonals
line
7.57
8
6
8
8
8
7
4
8
5
5
diagonals
heterosexual
0.11
0
1
0
0
0
3
0
0
0
0
naturalise
adapt
7.75
7
8
9
9
7
8
0
6
0
8
insubordinate
rebellious
9
10
8
8
9
10
8
9
10
9
9
insubordinate
defiant
9.56
10
9
9
10
10
7
10
10
9
9
subdividing
subdivide
9.5
9
10
10
10
10
6
10
9
7
8
subdividing
separate
8.67
9
9
8
8
10
6
10
9
9
0
antifeminism
sexism
8.6
10
9
9
6
10
5
5
9
9
7
circumcising
cut
8
7
8
8
7
8
10
10
8
8
7
circumcising
remove
8.17
8
8
6
6
8
10
9
8
8
5
excommunicate
oust
8.2
8
10
7
10
9
10
10
8
8
8
accomplished
over
5
7
0
4
0
10
10
0
7
2
5
attackers
wrongdoer
7.5
7
8
5
7
7
9
7
8
8
8
contravened
disagree
7.38
6
9
9
6
8
7
7
10
2
7
contravened
transgress
7
8
6
9
6
8
10
7
2
4
7
tenderize
change
5.88
7
0
4
6
4
5
7
7
7
0
blithering
chatter
7.71
8
10
6
7
9
10
6
9
4
9
resurfacing
coat
8.44
7
8
9
9
10
10
7
10
0
6
resurfacing
surface
6.5
8
5
6
4
8
5
8
7
10
5
friendships
brotherhood
7.5
8
7
6
8
7
6
9
7
9
8
soulfully
emotional
7.6
7
9
3
9
5
8
7
8
5
8
elector
voter
7.83
10
7
5
10
10
7
8
9
9
7
intentionality
intended
8.4
8
8
7
10
7
8
9
9
7
10
vulgarism
profanity
9.62
10
9
10
10
9
8
9
10
10
7
vulgarism
inelegance
6.29
5
4
6
9
1
9
8
8
7
6
preliterate
illiterate
4.71
5
7
5
5
5
7
3
5
5
0
preliterate
noncivilized
4.43
4
7
0
4
5
6
2
5
5
0
retrying
hear
0.38
0
4
0
0
3
5
0
0
0
0
wanderers
program
0.29
0
4
0
0
2
4
0
0
5
0
wanderers
nomad
9.5
10
9
10
10
10
9
9
9
10
9
marginalize
interact
2.5
3
0
0
2
5
0
6
0
3
2
hyperlink
link
9.12
8
10
9
4
6
10
8
8
10
10
inducted
admit
7.17
5
7
3
6
8
10
10
8
0
9
inducted
install
6
8
7
0
4
7
10
4
6
0
6
entrapping
capture
8
8
8
3
10
7
8
8
8
9
8
entrapping
deceive
6.57
5
8
4
8
7
3
9
2
7
7
alleviated
comfort
6
6
6
5
4
8
6
6
6
8
8
alleviated
help
6
6
6
5
7
6
7
4
3
8
5
radiators
beginning
0
0
0
0
5
4
0
0
0
0
0
radiators
system
1
0
2
0
6
6
1
7
0
4
0
postmodernism
genre
2.8
0
6
1
3
7
3
7
2
0
5
excavations
site
5.29
3
0
1
6
7
9
5
5
6
5
excavations
removal
7.43
3
6
0
8
8
10
9
8
5
8
comfortable
cozy
8.83
9
9
5
10
8
9
7
9
9
10
unfeathered
unfledged
7.14
6
6
9
9
4
8
8
3
0
10
unfeathered
plucked
8
10
8
9
9
7
7
9
7
10
6
assigned
regiment
3.8
1
6
0
7
6
0
2
0
4
8
assigned
allow
3.5
1
5
0
7
9
0
5
0
3
7
listeners
eavesdropper
8.14
8
6
5
7
10
8
9
8
8
9
unheralded
unexpected
7.57
3
6
5
9
8
9
10
8
8
3
inabilities
insufficiency
7.38
6
8
1
8
8
3
8
6
8
7
inabilities
incomprehension
4.33
2
7
3
6
0
6
4
0
5
7
monocultures
culture
4.5
8
5
5
8
1
4
6
3
8
4
tricolour
flag
5.8
7
6
6
3
0
1
10
7
0
9
omnipotence
state
0.86
5
3
0
6
2
1
0
0
6
0
mingles
change
1.71
2
5
1
2
0
1
2
1
0
3
calcify
harden
8.12
10
8
3
9
6
9
8
9
7
9
calcify
change
4.83
5
7
0
5
4
5
5
0
8
5
disinheritance
discontinuance
5.67
4
6
0
8
6
4
0
0
7
7
interwove
braid
8.75
9
9
4
9
9
8
10
8
9
9
cession
relinquishment
6.67
6
7
8
8
1
6
2
5
10
0
dwarfish
small
8.38
9
7
3
10
9
9
9
7
9
8
assessments
charge
5.12
7
4
3
6
0
6
8
7
5
3
assessments
classification
6.86
4
8
1
6
8
10
9
7
0
6
prescriptions
medicine
9.11
10
5
9
9
9
10
8
9
9
9
prescriptions
direction
2.57
5
1
7
5
1
0
6
1
4
1
End of preview. Expand in Data Studio

Stanford Rare Word (RW) Similarity Dataset

Created by Minh-Thang Luong, Richard Socher, and Christopher D. Manning, Stanford University Computer Science Department.

Available at: http://nlp.stanford.edu/~lmthang/morphoNLM

Described in: Luong, M.-T., Socher, R., & Manning, C. D. (2013). Better Word Representations with Recursive Neural Networks for Morphology. CoNLL, Sofia, Bulgaria.

Columns

  1. Word1 – First word in the pair
  2. Word2 – Second word in the pair
  3. Avg_Similarity – Average human similarity score (0–10) 4–13. Rating1 … Rating10 – Up to 10 individual ratings (some rows have fewer)

CSV Formatting

I did not create or own this dataset, only converted the original rw.txt into a columnized CSV / Parquet for easier use. All credit belongs to the original authors.
Downloads last month
34
Number of rows:
2,034
Total file size:
98.3 kB