Proceedings of the Second Joint Conference on Lexical and Computational Semantics
UMBC_EBIQUITY-CORE: Semantic Textual Similarity Systems
June 13, 2013
We describe three semantic text similarity systems developed for the *SEM 2013 STS shared task and the results of the corresponding three runs. All of them used a word similarity feature that combined LSA word similarity and WordNet knowledge. The first run, which achieved the top mean score on the task of all the submissions, used a simple term alignment algorithm. The other two runs, ranked second and fourth, used SVM models to combine a larger sets of features.
Association for Computational Linguistics
Downloads: 3714 downloads
- (Publication) UMBC_EBIQUITY-CORE: Semantic Textual Similarity Systems has resource (Resource) UMBC webbase corpus