Semantic Textual Similarity of Portuguese-Language Texts: An Approach Based on the Semantic Inferentialism Model

The Semantic Textual Similarity (STS) task aims capturing a bidirectional-graded equivalence between the pair of short texts. This work proposes a STS measure for the Portuguese Language based on Semantic Inferentialism Model (SIM) and InferenceNet.BR. We argue that the expression of inferential, causal, motivational and encyclopedic content of InferenceNet enables a more robust and efficient model for the STS task. An extrinsic evaluation in a Portuguese-language processing application - a Case-Based Reasoning system for Requirements Engineering - provided real scenario to assess how the proposed STS measure contributes to the effectiveness of NLP applications.