The introduction of embedding techniques has pushed forward significantly the Natural Language Processing field. Many of the proposed solutions have been presented for word-level encoding; anyhow, in the last years, new mechanisms to treat information at a higher level of aggregation, like at sentence- and document-level, have emerged. With this work, we address specifically the sentence embeddings problem, presenting the Static Fuzzy Bag-of-Word model. Our model is a refinement of the Fuzzy Bag-of-Words approach, providing sentence embeddings with a fixed dimension. SFBoW provides competitive performances in Semantic Textual Similarity benchmarks while requiring low computational resources.

Static Fuzzy Bag-of-Words: a Lightweight and Fast Sentence Embedding Algorithm

Matteo Muffo;Licia Sbattella;Roberto Tedesco;Vincenzo Scotti
2021-01-01

Abstract

The introduction of embedding techniques has pushed forward significantly the Natural Language Processing field. Many of the proposed solutions have been presented for word-level encoding; anyhow, in the last years, new mechanisms to treat information at a higher level of aggregation, like at sentence- and document-level, have emerged. With this work, we address specifically the sentence embeddings problem, presenting the Static Fuzzy Bag-of-Word model. Our model is a refinement of the Fuzzy Bag-of-Words approach, providing sentence embeddings with a fixed dimension. SFBoW provides competitive performances in Semantic Textual Similarity benchmarks while requiring low computational resources.
Proceedings of The Fourth International Conference on Natural Language and Speech Processing (ICNLSP 2021)
978-1-955917-18-6
Semantic Textual Similarity; Fuzzy Sets; Natural Language Processing; Sentence Embeddings
File in questo prodotto:
File Dimensione Formato  
Static Fuzzy Bag-of-Words a Lightweight and Fast.pdf

accesso aperto

: Post-Print (DRAFT o Author’s Accepted Manuscript-AAM)
Dimensione 272.16 kB
Formato Adobe PDF
272.16 kB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11311/1187308
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact