Monotonicity Analysis for Paraphrase Detection

In this paper, we analyse the role of monotonicity for the recognition of paraphrasing pairs. Our approach is based in a system architecture which consists of two components. The first component is the alignment module which takes care of the order of the elements for the analysis of monotonicity. The semantic analysis is carried out by the second module which relies on the combination of semantic information provided by the analysis of lexical relationships, and the use of semantic heuristics to recognize false paraphrasing. The results of the experimentation conducted show how lexical coupling without false monotonicity assumption leads to higher accuracy. Moreover, the results of the experimentation using monotonic alignment techniques suggest that the MSR corpus might be a data set without a substantial syntactic diversity.