On the Replicability of Combining Word Embeddings and Retrieval Models

We replicate recent experiments attempting to demonstrate an attractive hypothesis about the use of the Fisher kernel framework and mixture models for aggregating word embeddings towards document representations and the use of these representations in document classification, clustering, and retriev...

Descripción completa

Detalles Bibliográficos
Autores principales: Papariello, Luca, Bampoulidis, Alexandros, Lupu, Mihai
Formato: Online Artículo Texto
Lenguaje:English
Publicado: 2020
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7148082/
http://dx.doi.org/10.1007/978-3-030-45442-5_7
_version_ 1783520527641477120
author Papariello, Luca
Bampoulidis, Alexandros
Lupu, Mihai
author_facet Papariello, Luca
Bampoulidis, Alexandros
Lupu, Mihai
author_sort Papariello, Luca
collection PubMed
description We replicate recent experiments attempting to demonstrate an attractive hypothesis about the use of the Fisher kernel framework and mixture models for aggregating word embeddings towards document representations and the use of these representations in document classification, clustering, and retrieval. Specifically, the hypothesis was that the use of a mixture model of von Mises-Fisher (VMF) distributions instead of Gaussian distributions would be beneficial because of the focus on cosine distances of both VMF and the vector space model traditionally used in information retrieval. Previous experiments had validated this hypothesis. Our replication was not able to validate it, despite a large parameter scan space.
format Online
Article
Text
id pubmed-7148082
institution National Center for Biotechnology Information
language English
publishDate 2020
record_format MEDLINE/PubMed
spelling pubmed-71480822020-04-13 On the Replicability of Combining Word Embeddings and Retrieval Models Papariello, Luca Bampoulidis, Alexandros Lupu, Mihai Advances in Information Retrieval Article We replicate recent experiments attempting to demonstrate an attractive hypothesis about the use of the Fisher kernel framework and mixture models for aggregating word embeddings towards document representations and the use of these representations in document classification, clustering, and retrieval. Specifically, the hypothesis was that the use of a mixture model of von Mises-Fisher (VMF) distributions instead of Gaussian distributions would be beneficial because of the focus on cosine distances of both VMF and the vector space model traditionally used in information retrieval. Previous experiments had validated this hypothesis. Our replication was not able to validate it, despite a large parameter scan space. 2020-03-24 /pmc/articles/PMC7148082/ http://dx.doi.org/10.1007/978-3-030-45442-5_7 Text en © Springer Nature Switzerland AG 2020 This article is made available via the PMC Open Access Subset for unrestricted research re-use and secondary analysis in any form or by any means with acknowledgement of the original source. These permissions are granted for the duration of the World Health Organization (WHO) declaration of COVID-19 as a global pandemic.
spellingShingle Article
Papariello, Luca
Bampoulidis, Alexandros
Lupu, Mihai
On the Replicability of Combining Word Embeddings and Retrieval Models
title On the Replicability of Combining Word Embeddings and Retrieval Models
title_full On the Replicability of Combining Word Embeddings and Retrieval Models
title_fullStr On the Replicability of Combining Word Embeddings and Retrieval Models
title_full_unstemmed On the Replicability of Combining Word Embeddings and Retrieval Models
title_short On the Replicability of Combining Word Embeddings and Retrieval Models
title_sort on the replicability of combining word embeddings and retrieval models
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7148082/
http://dx.doi.org/10.1007/978-3-030-45442-5_7
work_keys_str_mv AT paparielloluca onthereplicabilityofcombiningwordembeddingsandretrievalmodels
AT bampoulidisalexandros onthereplicabilityofcombiningwordembeddingsandretrievalmodels
AT lupumihai onthereplicabilityofcombiningwordembeddingsandretrievalmodels