UMCC_DLSI: Textual Similarity based on Lexical-Semantic features

Second Joint Conference on Lexical and Computational Semantics (*SEM), Volume 1: Proceedings of the Main Conference and the Shared Task, pages 109–118, Atlanta, Georgia, June 13-14, 2013. c⃝2013 Association for Computational Linguistics UMCC_DLSI: Textual Similarity based on Lexical-Semantic features Alexander Chávez, Antonio Fernández Orquín, Héctor Dávila, Yoan Gutiérrez, Armando Collazo, José I. Abreu DI, University of Matanzas Autopista a Varadero km 3 ½ Matanzas, Cuba. {alexander.chavez, tony, hector.davila, yoan.gutierrez, armando.collazo, jose.abreu}@umcc.cu Andrés Montoyo, Rafael Muñoz DLSI, University of Alicante Carretera de San Vicente S/N Alicante, Spain. {montoyo,rafael}@dlsi.ua.es Abstract This paper describes the specifications and results of UMCC_DLSI system, which participated in the Semantic Textual Similarity task (STS) of SemEval-2013. Our supervised system uses different types of lexical and semantic features to train a Bagging classifier used to decide the correct option. Related to the different features we can highlight the resource ISR-WN used to extract semantic relations among words and the use of different algorithms to establish semantic and lexical similarities. In order to establish which features are the most appropriate to improve STS results we participated with three runs using different set of features. Our best run reached the position 44 in the official ranking, obtaining a general correlation coefficient of 0.61. 1 Introduction SemEval-2013 (Agirre et al., 2013) presents the task Semantic Textual Similarity (STS) again. In STS, the participating systems must examine the degree of semantic equivalence between two sentences. The goal of this task is to create a unified framework for the evaluation of semantic textual similarity modules and to characterize their impact on NLP applications. STS is related to Textual Entailment (TE) and Paraphrase tasks. The main difference is that STS assumes bidirectional graded equivalence between the pair of textual snippets. In case of TE, the equivalence is directional (e.g. a student is a person, but a person is not necessarily a student). In addition, STS differs from TE and Paraphrase in that, rather than being a binary yes/no decision, STS is a similarity-graded notion (e.g. a student is more similar to a person than a dog to a person). This graded bidirectional is useful for NLP tasks such as Machine Translation (MT), Information Extraction (IE), Question Answering (QA), and Summarization. Several semantic tasks could be added as modules in the STS framework, “such as Word Sense Disambiguation and Induction, Lexical Substitution, Semantic Role Labeling, Multiword Expression detection and handling, Anaphora and Co-reference resolution, Time and Date resolution and Named Entity, among others”1 1.1 Description of 2013 pilot task This edition of SemEval-2013 remain with the same classification approaches that in their first version in 2012. The output of different systems was compared to the reference scores provided by SemEval-2013 gold standard file, which range from five to zero according to the next criterions2: (5) “The two sentences are equivalent, as they mean the same thing”. (4) “The two sentences are mostly equivalent, but some unimportant details differ”. (3) “The two sentences are roughly equivalent, but some important information differs/missing”. (2) “The two sentences are …

Content truncated to 500 words