A Classifier-Based Approach to Predict the Approval of Legislative Propositions

Ilo Cabral, Glauco Pedrosa

2023

Abstract

This paper presents a data mining-based approach to predict the approval of Legislative Propositions (LPs) based on textual documents. We developed a framework using machine learning and natural language processing algorithms for automatic text classification to predict whether or not a proposition would be approved in the legislative houses based on previous legislative proposals. The major contribution of this work is a novel kNN-based classifier less sensitive to imbalanced data and a time-wise factor to weight similar documents that are distant in time. This temporal factor aims to penalize the approval of LPs with subjects that are far from current political, social and cultural trends. The results obtained show that the proposed classifier increased the F1-score by 30% when compared to other traditional classifiers, demonstrating the potential of the proposed framework to assist political agents in the legislative process.

Paper Citation

in Harvard Style

Cabral I. and Pedrosa G. (2023). A Classifier-Based Approach to Predict the Approval of Legislative Propositions. In Proceedings of the 25th International Conference on Enterprise Information Systems - Volume 1: ICEIS, ISBN 978-989-758-648-4, SciTePress, pages 436-442. DOI: 10.5220/0011728800003467

in Bibtex Style

@conference{iceis23,
author={Ilo Cabral and Glauco Pedrosa},
title={A Classifier-Based Approach to Predict the Approval of Legislative Propositions},
booktitle={Proceedings of the 25th International Conference on Enterprise Information Systems - Volume 1: ICEIS,},
year={2023},
pages={436-442},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011728800003467},
isbn={978-989-758-648-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 25th International Conference on Enterprise Information Systems - Volume 1: ICEIS,
TI - A Classifier-Based Approach to Predict the Approval of Legislative Propositions
SN - 978-989-758-648-4
AU - Cabral I.
AU - Pedrosa G.
PY - 2023
SP - 436
EP - 442
DO - 10.5220/0011728800003467
PB - SciTePress