Feature Selection for Monitoring Food Security Key Variables in the State of Ceará, Brazil

• Ícaro L. Rodrigues Federal University of Ceará (UFC) http://orcid.org/0000-0001-8915-2972
• Luiza C. A. Pacheco Federal University of Ceará (UFC)
• Josué M. Hinrichs Federal University of Ceará (UFC)
• Adilio J. Freitas Federal University of Ceará (UFC)
• José Luciano M. Neto Federal University of Ceará (UFC)
• Antonio Rafael Braga Federal University of Ceará (UFC)
• Danielo G. Gomes Federal University of Ceará (UFC)

DOI: https://doi.org/10.5753/bresci.2024.244299

Abstract

The Brazilian state of Ceará presented a detrimental context in its food insecurity (FI) rate in 2023, with 35% of the total population experiencing some level of FI. This study aims to identify a subset of the most indicative variables regarding FI in Ceará, with the objective of improving public policies to combat hunger in the state. For this purpose, data from the Food Security module of the Pesquisa Nacional por Amostra de Domicílios Contínua (PNADC) 2023 were used. Five feature selection techniques were applied to a set of pre-processed variables, and the 18 most frequent variables were selected, with Education and Income/Employment categories standing out.

Keywords: Food Security, Feature Selection, Data Science, Digital Government

References

Bezerra, T. A., Olinda, R. A. d., and Pedraza, D. F. (2017). Insegurança alimentar no brasil segundo diferentes cenários sociodemográficos. Ciência & Saúde Coletiva, 22:637.

Boas, L. G. V. (2023). A escala brasileira de insegurança alimentar (EBIA) e as principais condicionantes da (in) segurança alimentar no brasil. Geoconexões, 1(15):114–134.

Chandrashekar, G. and Sahin, F. (2014). A survey on feature selection methods. Computers & electrical engineering, 40(1):16–28.

Cherol, C. C. d. S., Ferreira, A. A., Lignani, J. d. B., and Salles-Costa, R. (2022). Regional and social inequalities in food insecurity in Brazil, 2013-2018. Cadernos de Saúde Pública, 38(12).

Feng, F., Li, K.-C., Yang, E., Zhou, Q., Han, L., Hussain, A., and Cai, M. (2023). A novel oversampling and feature selection hybrid algorithm for imbalanced data classification. Multimedia Tools and Applications, 82(3):3231–3267.

Gandra, A. (2022). 2º inquérito nacional sobre insegurança alimentar no contexto da pandemia da covid-19 no brasil: Pesquisa aponta que fome atinge 33,1 milhões de pessoas no país. Agência Brasil, Rio de Janeiro, 8:2022–06.

Golgher, A. B. (2024). Food insecurity in brazil by household arrangements and characteristics between 2004 and 2022. Cadernos de Saúde Pública, 40(5).

Gomes, I. (2023). Pobreza cai para 31,6% da população em 2022, após alcançar 36,7% em 2021. [link]. Acesso em: 22 de julho 2024.

Gosain, A. and Sardana, S. (2017). Handling class imbalance problem using oversampling techniques: A review. In 2017 international conference on advances in computing, communications and informatics (ICACCI), pages 79–85. IEEE.

Lemaître, G., Nogueira, F., and Aridas, C. K. (2017). Imbalanced-learn: A python toolbox to tackle the curse of imbalanced datasets in machine learning. Journal of Machine Learning Research, 18(17):1–5.

Menardi, G. and Torelli, N. (2014). Training and assessing classification rules with imbalanced data. Data mining and knowledge discovery, 28:92–122.

PENSSAN and II VIGISAN (2021). Insegurança alimentar e covid-19 no brasil: inquérito nacional sobre insegurança alimentar no contexto da pandemia da covid-19 no brasil. Belo Horizonte: Instituto Vox Populi.

Zhang, C., Soda, P., Bi, J., Fan, G., Almpanidis, G., Garc´ıa, S., and Ding, W. (2023). An empirical study on the joint impact of feature selection and data resampling on imbalance classification. Applied Intelligence, 53(5):5449–5461.