Skip to Main Content
Table 4 

Detailed results as F1 for the four data sets studied with tf-idf and mdice as measures. The results show the performance of our unsupervised (uns) and semi-supervised (ssup) system and the results obtained using the most frequent sense heuristic (MFS). Detailed information about the performance of the systems on different parts of speech are provided: nouns (N), verbs (V), adjectives (A), and adverbs (R).

SemEval 2007 coarse-grained - S7CG
MethodAllNVAR
WSDgamesuns 80.4 85.5 71.2 81.5 76.0 
WSDgamesssup 82.8 85.4 77.2 82.9 84.6 
MFS 76.3 76.0 70.1 82.0 86.0 
 
SemEval 2007 fine-grained - S7 
Method All 
WSDgamesuns 43.3 49.7 39.9 − − 
WSDgamesssup 56.5 62.9 53.0 − − 
MFS 54.7 60.4 51.7 − − 
 
Senseval 3 fine-grained - S3 
Method All 
WSDgamesuns 59.1 63.3 50.7 64.5 71.4 
WSDgamesssup 64.7 70.3 54.1 70.7 85.7 
MFS 62.8 69.3 51.4 68.2 100.0 
 
Senseval 2 fine-grained - S2 
Method All 
WSDgamesuns 61.2 69.8 41.7 61.9 65.1 
WSDgamesssup 66.0 72.4 43.5 71.8 75.7 
MFS 65.6 72.1 42.4 71.6 76.1 
SemEval 2007 coarse-grained - S7CG
MethodAllNVAR
WSDgamesuns 80.4 85.5 71.2 81.5 76.0 
WSDgamesssup 82.8 85.4 77.2 82.9 84.6 
MFS 76.3 76.0 70.1 82.0 86.0 
 
SemEval 2007 fine-grained - S7 
Method All 
WSDgamesuns 43.3 49.7 39.9 − − 
WSDgamesssup 56.5 62.9 53.0 − − 
MFS 54.7 60.4 51.7 − − 
 
Senseval 3 fine-grained - S3 
Method All 
WSDgamesuns 59.1 63.3 50.7 64.5 71.4 
WSDgamesssup 64.7 70.3 54.1 70.7 85.7 
MFS 62.8 69.3 51.4 68.2 100.0 
 
Senseval 2 fine-grained - S2 
Method All 
WSDgamesuns 61.2 69.8 41.7 61.9 65.1 
WSDgamesssup 66.0 72.4 43.5 71.8 75.7 
MFS 65.6 72.1 42.4 71.6 76.1 
Close Modal

or Create an Account

Close Modal
Close Modal