Paper Information | SBC BOOKS

Fill in your paper information

Language (*)

Title (*)

Keywords (*)

Abstract (*)

The growing demand for scholarships in higher education and the limited financial resources make it challenging to select candidates with the most significant socioeconomic vulnerability. To assist in this task, we developed three graph-based models to represent the network of candidates for student aid. We applied the Community Detection methods Fast Greedy, Multilevel, and Walktrap to the networks generated by each model to identify groups of candidates with the most significant similarity in socioeconomic characteristics. Computational experiments were performed with real data from three years of aid applications from a Brazilian federal university to validate the models. We analyzed the generated solutions for cohesion and external isolation between communities based on metrics such as modularity and the number of edges between communities. Model 2 (Multirelational) obtained the best results in the three datasets analyzed, with the Fast Greedy method delivering more cohesive communities than the others. We noticed that the Community Detection method greatly influences the obtaining of better quality solutions, which in some cases reduced the inter-community edges by about eight times in cases with the same number of communities. The work brings contributions by presenting three network models for academic management, a context that still needs to be explored in the literature, which, together with Community Detection methods, can potentially classify data in an unsupervised manner.

Pages (*)

File Link

English Information

Title

Keywords

Abstract

(*) To change the order drag the item to the new position.

Authors

#	Name
1	Rodrigo Porto(rodrigoandradeporto@hotmail.com)
2	Rafael Frinhani(frinhani@unifei.edu.br)
3	Felipe Mendes(felipe.mendes@unifei.edu.br)
4	Vanessa Souza(vanessasouza@unifei.edu.br)

(*) To change the order drag the item to the new position.

Reference

#	Reference
1	Barabási, A.-L. and Pósfai, M. Network science. Cambridge University Press, Cambridge, 2016.
2	Bassi, F. and Vera, José Fernando nd Martín, J. A. M. Profile-based latent class distance association analyses for sparse tables: application to the attitude of european citizens towards sustainable tourism. Advances in Data Analysis and Classification vol. 18, pp. 1–28, 2023.
3	Diboune, A., Slimani, H., Nacer, H., and Beghdad Bey, K. A comprehensive survey on community detection methods and applications in complex information networks. Social Network Analysis and Mining 14 (1): 1–47, 2024.
4	Fortunato, S. Community detection in graphs. Physics Reports 486 (3): 75–174, 2010.
5	Hamim, T., Benabbou, F., and Sael, N. Survey of machine learning techniques for student profile modeling. International Journal of Emerging Technologies in Learning (iJET) 16 (04): 136–151, 2021.
6	Newman, M. E. and Girvan, M. Finding and evaluating community structure in networks. Physical review E 69 (2): 026113, 2004.
7	Ramesh, A., Rodriguez, M., and Getoor, L. Multi-relational influence models for online professional networks. In Proceedings of the International Conference on Web Intelligence. WI ’17. Association for Computing Machinery, New York, NY, USA, pp. 291–298, 2017.
8	Yang, H.-W., Pan, Z.-G., Wang, X.-Z., and Xu, B. A personalized products selection assistance based on e-commerce machine learning. In Proceedings of 2004 International Conference on Machine Learning and Cybernetics (ICMLC). Vol. 4. IEEE, Shanghai, China, pp. 2629–2633, 2004.
9	Yang, Z., Algesheimer, R., and Tessone, C. J. A comparative analysis of community detection algorithms on artificial networks. Nature Scientific Reports vol. 1, pp. 1–16, 2016.

Paper Registration

Fill in your paper information

English Information

Authors

Reference