Papers by Boško Nikolić
Fine-grained Semantic Textual Similarity for Serbian (L18-1)
Copied to clipboard
| Challenge: | Semantic textual similarity (STS) is a task of assigning a numerical score to short texts based on the level of semantic equivalence between them. |
| Approach: | They propose to annotate Serbian STS dataset with fine-grained similarity scores . they propose a supervised bag-of-words model that combines part-of speech weighting with term frequency weighting . |
| Outcome: | The proposed model outperforms existing models on the Serbian STS News Corpus . the proposed model is based on a new morphologically rich language . |