Measuring short Text Semantic Similarity using multiple measurements

In this paper, we present a Support Vector Regression (SVR) system to measure the semantic similarity of short texts by combining multiple similarity measurements, i.e., string similarity, knowledge-based similarity, corpus-based similarity, syntactic dependency similarity, number similarity and machine translation similarity. Experiments on the five data sets of SemEval 2012 Semantic Text Similarity (STS) task show that our system performs best on two data sets, and second best on another two data sets.