Code Smell Classification in Python: Are Small Language Models Up to the Task?

Gespeichert in:
Bibliographische Detailangaben
Titel: Code Smell Classification in Python: Are Small Language Models Up to the Task?
Autoren: Soares de Oliveira, Igor, orcid:0009-0002-4554-, Carneiro da Silva Ribeiro, Joanne, Ribas, Jessica, Alves Pereira, Juliana
Verlagsinformationen: Zenodo
Publikationsjahr: 2025
Bestand: Zenodo
Beschreibung: This repository brings together the artifacts developed as part of a study focused on the classification of code smells using Small Language Models (SLMs). It consists of a set of applications and a dataset. Applications: Three services that operate in an integrated manner to perform the extraction, analysis, and storage of information related to code smells found in code snippets. These services include a submission interface, mechanisms for structured code extraction, and integration with language models for the classification of smells. Dataset: A dataset generated from the experiments conducted, containing examples of code smells classified through the use of SLMs.
Publikationsart: software
Sprache: unknown
Relation: https://zenodo.org/records/16999902; oai:zenodo.org:16999902; https://doi.org/10.5281/zenodo.16999902
DOI: 10.5281/zenodo.16999902
Verfügbarkeit: https://doi.org/10.5281/zenodo.16999902
https://zenodo.org/records/16999902
Rights: Creative Commons Attribution 4.0 International ; cc-by-4.0 ; https://creativecommons.org/licenses/by/4.0/legalcode
Dokumentencode: edsbas.80C49CFE
Datenbank: BASE
Beschreibung
Abstract:This repository brings together the artifacts developed as part of a study focused on the classification of code smells using Small Language Models (SLMs). It consists of a set of applications and a dataset. Applications: Three services that operate in an integrated manner to perform the extraction, analysis, and storage of information related to code smells found in code snippets. These services include a submission interface, mechanisms for structured code extraction, and integration with language models for the classification of smells. Dataset: A dataset generated from the experiments conducted, containing examples of code smells classified through the use of SLMs.
DOI:10.5281/zenodo.16999902