ExB Themis: Extensive Feature Extraction from Word Alignments for Semantic Textual Similarity

摘要

We present ExB Themis - a word alignment-based semantic textual similarity system developed for SemEval-2015 Task 2: Semantic Textual Similarity. It combines both string and semantic similarity measures as well as alignment features using Support Vector Regression. It occupies the first three places on Spanish data and additionally places second on English data. ExB Themis proved to be the best multilingual system among all participants.