SOFTCARDINALITY-CORE: Improving Text Overlap with Distributional Measures for Semantic Textual Similarity

The soft cardinality proved to be a very strong text-overlapping baseline for the task of semantic-textual-similarity (STS) obtaining the third place in SemEval-2012. This year, besides to the plain text-overlapping approach, two distributional word-similarity functions derived from the ukWack corpus were tested within the soft cardinality. These measures contributed to improve the performance of the text-overlapping approach. Further, these were combined with other features using regression obtaining positions 18th, 22th and 23th among the 90 participants systems in the official 2013 shared task ranking at *SEM. After the release of the gold standard anotations of the test data, we observed that the bare similarity measures, without the use of regression, would have obtained positions 6th, 7th and 8th. Moreover, the simple arithmetic average of these similarity measures would have been 4th (mean=0.5747). This paper describes the submitted system and the similarity measures that would obtained those better results.