Identificação de Características Relevantes para Construir Modelos Preditivos de Desempenhos de Alunos
DOI:
https://doi.org/10.22491/1982-1654.135845Resumo
Este trabalho identificou um subconjunto de características relevantes em um conjunto de dados público para cada algoritmo de aprendizagem de máquina, e classificou os desempenhos dos alunos com maior exatidão. O conjunto de dados contém informações sobre notas, faltas, características demográficas e sociais dos alunos. A metodologia proposta usou o algoritmo Recursive Feature Elimination (RFE) para seleção de características e algoritmos de aprendizagem de máquina baseados nas abordagens de classificação Decision Tree, Gradient Tree Boosting e Support Vector Machines (SVM). O modelo construído do SVM alcançou a maior média da macro F1 na validação cruzada, 0,8350 na pontuação macro F1 do treino; na fase de teste, alcançou 0,8365 na pontuação macro F1. Esse modelo foi construído com quatorze atributos selecionados pelo RFE. A análise dos resultados constatou que a identificação das características mais relevantes para predizer os desempenhos dos alunos depende da abordagem de aprendizagem de cada modelo.
Downloads
Referências
Abdullahi, Wokili, Mary Ogbuka Kenneth e Morufu Olalere (2021). “Student perfor- mance prediction using a cascaded bi-level feature selection approach.” Em: Jour- nal of Computer Science Research 3.3. GS Search., pp. 16–28.
Ahmed, Saja Taha, Rafah Al-Hamdani e Muayad Sadik Croock (2020). “Developed third iterative dichotomizer based on feature decisive values for educational data mi- ning”. Em: Indonesian Journal of Electrical Engineering and Computer Science
1. GS Search, pp. 209–217.
Ajibade, Samuel-Soma M., Nor Bahiah Binti Ahmad e Anazida Zainal (2020). “Analy- sis of metaheuristics feature selection algorithm for classification”. Em: Hybrid Intelligent Systems - 20th International Conference on Hybrid Intelligent Systems (HIS 2020), Virtual Event, India, December 14-16, 2020. Ed. por Ajith Abraham, Thomas Hanne, Oscar Castillo, Niketa Gandhi, Tatiane Nogueira Rios e Tzung- Pei Hong. Vol. 1375. Advances in Intelligent Systems and Computing. Springer, pp. 214–222. URL: https : / / link . springer . com / chapter / 10 . 1007/978-3-030-73050-5_21.
Alamri, Leena H., Ranim S. Almuslim, Mona S. Alotibi, Dana K. Alkadi, Irfan Ullah
Khan e Nida Aslam (2021). “Predicting student academic performance using sup- port vector machine and random forest”. Em: Proceedings of the 2020 3rd Inter- national Conference on Education Technology Management. ICETM ’20. GS Se- arch. London, United Kingdom: Association for Computing Machinery, pp. 100– 107. ISBN: 9781450388757. DOI: 10.1145/3446590.3446607.
Albreiki, Balqis, Nazar Zaki e Hany Alashwal (2021). “A systematic literature review of
student’ performance prediction using machine learning techniques”. Em: Educa- tion Sciences 11.9. GS Search, pp. 1–27.
Baeza-Yates, Ricardo e Berthier Ribeiro-Neto (2011). Modern information retrieval: the concepts and technology behind search. 2ª ed. GS Search. Boston, MA: Addison- Wesley Publishing Company.
Baker, Ryan, Seiji Isotani e Adriana Carvalho ( de 2011). “Minerac¸a˜o de dados educa- cionais: oportunidades para o Brasil”. Em: Revista Brasileira de Informa´tica na Educac¸a˜o 19.02. GS Search, p. 00. ISSN: 1414-5685.
Buitinck, Lars, Gilles Louppe, Mathieu Blondel, Fabian Pedregosa, Andreas Mueller, Oli- vier Grisel, Vlad Niculae, Peter Prettenhofer, Alexandre Gramfort, Jaques Gro- bler, Robert Layton, Jake VanderPlas, Arnaud Joly, Brian Holt e Gae¨l Varoquaux
(2013). “API design for machine learning software: experiences from the scikit- learn project”. Em: CoRR abs/1309.0238. arXiv: 1309.0238. URL: http:// arxiv.org/abs/1309.0238.
Chen, Tianqi e Carlos Guestrin (2016). “XGBoost: A Scalable Tree Boosting System”. Em: CoRR abs/1603.02754. arXiv: 1603 . 02754. URL: http : / / arxiv . org/abs/1603.02754.
Chettaoui, Neila, Ayman Atia e Med Salim Bouhlel (2022). “Predicting students per- formance using eye-gaze features in an embodied Learning Environment”. Em: IEEE Global Engineering Education Conference, EDUCON 2022, Tunis, Tunisia, March 28-31, 2022. Ed. por Ilhem Kallel, Habib M. Kammoun e Lobna Hsairi. IEEE, pp. 704–711. DOI: 10.1109/EDUCON52537.2022.9766783. URL:
https://doi.org/10.1109/EDUCON52537.2022.9766783.
Costa, Rebeca Soler, Qing Tan, Fre´de´rique Pivot, Xiaokun Zhang e Harris Wang (2021). “Personalized and adaptive learning: educational practice and technological im- pact”. Em: Texto Livre 14.3. GS Search, e33445. ISSN: 1983-3652.
Dabhade, Pranav, Ravina Agarwal, K.P. Alameen, A.T. Fathima, R. Sridharan e G. Go- pakumar (2021). “Educational data mining for predicting students’ academic per- formance using machine learning algorithms”. Em: Materials Today: Proceedings
GS Search, pp. 5260–5267. ISSN: 2214-7853. DOI: https://doi.org/ 10.1016/j.matpr.2021.05.646.
Dua, Dheeru e Casey Graff (2017). UCI Machine learning repository. URL: http:// archive.ics.uci.edu/ml.
Freitas, Alex A (2002). Data mining and knowledge discovery with evolutionary algo- rithms. GS Search. Springer Science & Business Media.
Guyon, Isabelle, Jason Weston, Stephen Barnhill e Vladimir Vapnik (2002). “Gene selec- tion for cancer classification using support vector machines”. Em: Machine lear- ning 46. GS Search, pp. 389–422. URL: https://link.springer.com/ article/10.1023/A:1012487302797.
Hengpraprohm, Kairung, Supoj Hengpraprohm e Wannee Sudjitjoon (2022). “A study
of factors affecting learning efficiency on higher education student performance evaluation dataset using feature selection techniques”. Em: Information Techno- logy Journal 18.2, pp. 34–43. URL: https://ph01.tci- thaijo.org/ index.php/IT_Journal/article/view/251051.
Hussain, Shah e Muhammad Qasim Khan (2021). “Student-performulator: predicting
students’ academic performance at secondary and intermediate level using ma- chine learning”. Em: Annals of data science 10.3. GS Search, pp. 637–655. URL: https://link.springer.com/article/10.1007/s40745-021- 00341-0.
Kabathova, Janka e Martin Drlik (2021). “Towards predicting student’s dropout in uni-
versity courses using different machine learning techniques”. Em: Applied Scien- ces 11.7. GS Search. ISSN: 2076-3417. DOI: 10.3390/app11073130. URL:
https://www.mdpi.com/2076-3417/11/7/3130.
Kampff, Adriana Justin Cerveira (mar. de 2010). “Minerac¸a˜o de Dados Educacionais para Gerac¸a˜o de Alertas em Ambientes Virtuais de Aprendizagem como Apoio a` Pra´tica Docente”. Em: Informa´tica na educac¸a˜o: teoria amp; pra´tica 12.2. DOI:
22456/1982- 1654.12310. URL: https://seer.ufrgs.br/ index.php/InfEducTeoriaPratica/article/view/12310.
Khanal, Shristi Shakya, PWC Prasad, Abeer Alsadoon e Angelika Maag (2020). “A sys- tematic review: machine learning based recommendation systems for e-learning”. Em: Education and Information Technologies 25.4. GS Search, pp. 2635–2664.
Kour, Sunpreet, Rakesh Kumar e Meenu Gupta (2021). “Analysis of student performance using Machine learning Algorithms”. Em: 2021 Third International Conference on Inventive Research in Computing Applications (ICIRCA). GS Search. IEEE, pp. 1395–1403.
Lemay, David John, Clare Baek e Tenzin Doleck (2021). “Comparison of learning analy- tics and educational data mining: a topic modeling approach”. Em: Computers and Education: Artificial Intelligence 2. GS Search, p. 100016. ISSN: 2666-920X.
DOI: https://doi.org/10.1016/j.caeai.2021.100016.
Manha˜es, Laci Mary Barbosa e SMS Cruz (2020). “Predic¸a˜o do desempenho acadeˆmico de alunos da graduac¸a˜o utilizando minerac¸a˜o de dados.” Em: Simpo´sio de Pes- quisa Operacional e Log´ıstica da Marinha - Publicac¸a˜o Online 3.1, pp. 2050– 2064. ISSN: 2175-6295. DOI: http : / / dx . doi . org / 10 . 5151 / spolm2019 - 148. URL: www . proceedings . blucher . com . br / article-details/34563.
Nidhi, Nidhi, Sachin Majithia e Neeraj Sharma (2021). “Predictive model for students’ academic performance using classification and feature selection techniques”. Em: 2021 2nd International Conference on Computational Methods in Science & Te- chnology (ICCMST). GS Search. IEEE, pp. 106–111.
Romero, Cristo´bal, Manuel-Ignacio Lo´pez, Jose-Mar´ıa Luna e Sebastia´n Ventura (2013). “Predicting students’ final performance from participation in on-line discussion forums”. Em: Computers Education 68, pp. 458–472. ISSN: 0360-1315. DOI: https : / / doi . org / 10 . 1016 / j . compedu . 2013 . 06 . 009. URL: https : / / www . sciencedirect . com / science / article / pii / S0360131513001607.
Santos, Fa´bio Paula (abr. de 2017). “Um modelo computacional de apoio a` ana´lise da
opinia˜o de alunos sobre pra´ticas docentes por meio da minerac¸a˜o de dados educa- cionais”. GS Search. Tese de dout. Sa˜o Paulo, SP, BR: Universidade Presbiteriana Mackenzie.
Santos, Fabricia Damando (abr. de 2016). “Descoberta do desaˆnimo de alunos em am- bientes virtuais de ensino e aprendizagem: um modelo a partir da minerac¸a˜o de dados educacionais”. GS Search. Tese de dout. Porto Alegre, RS, BR: Universi- dade Federal do Rio Grande do Sul - UFRGS.
Sekeroglu, Boran, Kamil Dimililer e Kubra Tuncal (2019). “Student performance predic- tion and classification using machine learning algorithms”. Em: Proceedings of the 2019 8th International Conference on Educational and Information Technology. ICEIT 2019. GS Search. Cambridge, United Kingdom: Association for Compu- ting Machinery, pp. 7–11. ISBN: 9781450362672. DOI: 10.1145/3318396. 3318419. URL: https://doi.org/10.1145/3318396.3318419.
Silva, Douglas H, Erick G Maziero, Muhammad Saadi, Renata L Rosa, Juan C Silva,
Demostenes Z Rodriguez e Kostromitin K Igorevich (2022). “Big data analy- tics for critical information classification in online social networks using clas-
sifier chains”. Em: Peer-to-Peer Networking and Applications 15.1. GS Search, pp. 626–641. URL: https : / / link . springer . com / article / 10 . 1007/s12083-021-01269-1.
Souza, Vanessa Faria de (2021). “Minerac¸a˜o de dados educacionais com aprendizagem de ma´quina”. Em: Revista Educar Mais 5.4. GS Search, pp. 766–787.
Tan, Pang-Ning, Michael Steinbach e Vipin Kumar (2005). Introduction to data mining, (First Edition). GS Search. Boston, MA, USA: Addison-Wesley Longman Pu- blishing Co., Inc. ISBN: 0321321367.
Vijayalakshmi, V e K Venkatachalapathy (2019). “Comparison of predicting student’s performance using machine learning algorithms”. Em: International Journal of Intelligent Systems and Applications 11.12. GS Search, pp. 34–45.
Witten, Ian H., Eibe Frank e Mark A. Hall (2011). Data mining: practical ma- chine learning tools and techniques, 3rd Edition. Morgan Kaufmann, Elsevier. ISBN: 9780123748560. URL: https : / / www . worldcat . org / oclc / 262433473.
Yakubu, Mohammed Nasiru e Abubakar Mohammed Abubakar (abr. de 2022). “Applying
machine learning approach to predict students’ performance in higher educational institutions”. Em: Kybernetes 51.2. GS Search, pp. 916–934. DOI: 10.1108/K- 12- 2020- 0865. URL: https://doi.org/10.1108/K- 12- 2020-
Yufeia, Liu, Salmiza Salehb, Huang Jiahuic e Syed Mohamad Syed (2020). “Review of the application of artificial intelligence in education”. Em: Integration (Amster- dam) 12.8. GS Search, pp. 1–15.
Downloads
Publicado
Como Citar
Edição
Seção
Licença
Este trabalho está licenciado sob uma licença Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Os direitos autorais para artigos publicados nesta revista são do autor, com direitos de primeira publicação para a revista. Em virtude de aparecerem nesta revista de acesso público, os artigos são de uso gratuito, com atribuições próprias, em aplicações educacionais e não-comerciais. Os textos podem ser compartilhados desde que respeitados os direitos autorais sob a licença Creative Commons Não comercial-CompartilhaIgual 4.0.Aceito 2024-05-20
Publicado 2024-06-30
