Paper: Learning Semantic Textual Similarity with Structural Representations

ACL ID P13-2125
Title Learning Semantic Textual Similarity with Structural Representations
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013
Authors

Measuring semantic textual similarity (STS) is at the cornerstone of many NLP applications. Different from the major- ity of approaches, where a large number of pairwise similarity features are used to represent a text pair, our model features the following: (i) it directly encodes input texts into relational syntactic structures; (ii) relies on tree kernels to handle feature engineering automatically; (iii) combines both structural and feature vector repre- sentations in a single scoring model, i.e., in Support Vector Regression (SVR); and (iv) delivers significant improvement over the best STS systems.