Alert button

SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition

May 14, 2019
Hongming Zhang, Hantian Ding, Yangqiu Song

Figure 1 for SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition
Figure 2 for SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition
Figure 3 for SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition
Figure 4 for SP-10K: A Large-scale Evaluation Set for Selectional Preference Acquisition

Share this with someone who'll enjoy it:

Selectional Preference (SP) is a commonly observed language phenomenon and proved to be useful in many natural language processing tasks. To provide a better evaluation method for SP models, we introduce SP-10K, a large-scale evaluation set that provides human ratings for the plausibility of 10,000 SP pairs over five SP relations, covering 2,500 most frequent verbs, nouns, and adjectives in American English. Three representative SP acquisition methods based on pseudo-disambiguation are evaluated with SP-10K. To demonstrate the importance of our dataset, we investigate the relationship between SP-10K and the commonsense knowledge in ConceptNet5 and show the potential of using SP to represent the commonsense knowledge. We also use the Winograd Schema Challenge to prove that the proposed new SP relations are essential for the hard pronoun coreference resolution problem.

* Accepted by ACL 2019  
View paper onarxiv icon

Share this with someone who'll enjoy it: