Detecting Paraphrases for Portuguese using Word and Sentence Embeddings
Abstract
Paraphrase detection/identification is the task of determining whether two or more sentences of arbitrary length possess the same meaning. Methods to solve this task have many potential applications in Natural Language Processing systems. This work investigates the combination of different methods of sentence representation in a vector space model of language and linear classifiers to the problem of paraphrase identification for the Portuguese language. The results obtained in this work are inferior to those obtained for the related task of recognizing textual entailment in the ASSIN evaluation for the Portuguese language, but we point out that in this work we investigate the application of sentence embeddings to the problem of paraphrase detection, as such other features usually explored in systems for this task may be trivially incorporated into our method to improve performance.
Copyright (c) 2019 Marlo Souza
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
