Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Moraes, Daniel de S., Santos, Pedro T. C., da Costa, Polyana B., Pinto, Matheus A. S., Pinto, Ivan de J. P., da Veiga, Álvaro M. G., Colcher, Sergio, Busson, Antonio J. G., Rocha, Rafael H., Gaio, Rennan, Miceli, Rafael, Tourinho, Gabriela, Rabaioli, Marcos, Santos, Leandro, Marques, Fellipe, Favaro, David
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2401.06790
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910325421375488
author Moraes, Daniel de S.
Santos, Pedro T. C.
da Costa, Polyana B.
Pinto, Matheus A. S.
Pinto, Ivan de J. P.
da Veiga, Álvaro M. G.
Colcher, Sergio
Busson, Antonio J. G.
Rocha, Rafael H.
Gaio, Rennan
Miceli, Rafael
Tourinho, Gabriela
Rabaioli, Marcos
Santos, Leandro
Marques, Fellipe
Favaro, David
author_facet Moraes, Daniel de S.
Santos, Pedro T. C.
da Costa, Polyana B.
Pinto, Matheus A. S.
Pinto, Ivan de J. P.
da Veiga, Álvaro M. G.
Colcher, Sergio
Busson, Antonio J. G.
Rocha, Rafael H.
Gaio, Rennan
Miceli, Rafael
Tourinho, Gabriela
Rabaioli, Marcos
Santos, Leandro
Marques, Fellipe
Favaro, David
contents This work presents an unsupervised method for automatically constructing and expanding topic taxonomies using instruction-based fine-tuned LLMs (Large Language Models). We apply topic modeling and keyword extraction techniques to create initial topic taxonomies and LLMs to post-process the resulting terms and create a hierarchy. To expand an existing taxonomy with new terms, we use zero-shot prompting to find out where to add new nodes, which, to our knowledge, is the first work to present such an approach to taxonomy tasks. We use the resulting taxonomies to assign tags that characterize merchants from a retail bank dataset. To evaluate our work, we asked 12 volunteers to answer a two-part form in which we first assessed the quality of the taxonomies created and then the tags assigned to merchants based on that taxonomy. The evaluation revealed a coherence rate exceeding 90% for the chosen taxonomies. The taxonomies' expansion with LLMs also showed exciting results for parent node prediction, with an f1-score above 70% in our taxonomies.
format Preprint
id arxiv_https___arxiv_org_abs_2401_06790
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Using Zero-shot Prompting in the Automatic Creation and Expansion of Topic Taxonomies for Tagging Retail Banking Transactions
Moraes, Daniel de S.
Santos, Pedro T. C.
da Costa, Polyana B.
Pinto, Matheus A. S.
Pinto, Ivan de J. P.
da Veiga, Álvaro M. G.
Colcher, Sergio
Busson, Antonio J. G.
Rocha, Rafael H.
Gaio, Rennan
Miceli, Rafael
Tourinho, Gabriela
Rabaioli, Marcos
Santos, Leandro
Marques, Fellipe
Favaro, David
Computation and Language
Artificial Intelligence
This work presents an unsupervised method for automatically constructing and expanding topic taxonomies using instruction-based fine-tuned LLMs (Large Language Models). We apply topic modeling and keyword extraction techniques to create initial topic taxonomies and LLMs to post-process the resulting terms and create a hierarchy. To expand an existing taxonomy with new terms, we use zero-shot prompting to find out where to add new nodes, which, to our knowledge, is the first work to present such an approach to taxonomy tasks. We use the resulting taxonomies to assign tags that characterize merchants from a retail bank dataset. To evaluate our work, we asked 12 volunteers to answer a two-part form in which we first assessed the quality of the taxonomies created and then the tags assigned to merchants based on that taxonomy. The evaluation revealed a coherence rate exceeding 90% for the chosen taxonomies. The taxonomies' expansion with LLMs also showed exciting results for parent node prediction, with an f1-score above 70% in our taxonomies.
title Using Zero-shot Prompting in the Automatic Creation and Expansion of Topic Taxonomies for Tagging Retail Banking Transactions
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2401.06790