Adaptive multilevel clustering model for the prediction of academic risk

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The selection of a model for academic risk prediction systems is usually based on the global performance of the model. However, this global performance is not an important factor for the end-user of the system. For the end-user, the performance of the model for his or her specific case is the most important aspect of that model. Given that the model is usually selected at design time, the end-user could end up with a sub-optimal prediction. To solve this problem, this work presents a conceptual framework to build adaptive multilevel clustering models for academic risk prediction. This frameworks allows the system to automatically select between several levels of hierarchical or semi-hierarchical features to create a clustering model to best predict the particular risk of each student. This conceptual framework is validated through its realization into an adaptive model to predict the risk of failing a course during a semester in a Computer Science program. In this study, the adaptive model consistently outperforms the prediction of the best-performing static model.

Original languageEnglish (US)
Title of host publicationProceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016
EditorsIsaac Alpizar Chacon, Mario Chacon Rivas, Cristian Cechinel, Agustin Francesa Alfaro
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509061495
DOIs
StatePublished - Nov 21 2016
Event11th Latin American Conference on Learning Objects and Technology, LACLO 2016 - San Carlos, Costa Rica
Duration: Oct 3 2016Oct 7 2016

Other

Other11th Latin American Conference on Learning Objects and Technology, LACLO 2016
CountryCosta Rica
CitySan Carlos
Period10/3/1610/7/16

Fingerprint

Cluster Analysis
Software
Students
performance
computer science
Computer science
semester

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Developmental and Educational Psychology
  • Education

Cite this

Ochoa, X. (2016). Adaptive multilevel clustering model for the prediction of academic risk. In I. A. Chacon, M. C. Rivas, C. Cechinel, & A. F. Alfaro (Eds.), Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016 [7751800] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LACLO.2016.7751800

Adaptive multilevel clustering model for the prediction of academic risk. / Ochoa, Xavier.

Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016. ed. / Isaac Alpizar Chacon; Mario Chacon Rivas; Cristian Cechinel; Agustin Francesa Alfaro. Institute of Electrical and Electronics Engineers Inc., 2016. 7751800.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ochoa, X 2016, Adaptive multilevel clustering model for the prediction of academic risk. in IA Chacon, MC Rivas, C Cechinel & AF Alfaro (eds), Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016., 7751800, Institute of Electrical and Electronics Engineers Inc., 11th Latin American Conference on Learning Objects and Technology, LACLO 2016, San Carlos, Costa Rica, 10/3/16. https://doi.org/10.1109/LACLO.2016.7751800
Ochoa X. Adaptive multilevel clustering model for the prediction of academic risk. In Chacon IA, Rivas MC, Cechinel C, Alfaro AF, editors, Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016. Institute of Electrical and Electronics Engineers Inc. 2016. 7751800 https://doi.org/10.1109/LACLO.2016.7751800
Ochoa, Xavier. / Adaptive multilevel clustering model for the prediction of academic risk. Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016. editor / Isaac Alpizar Chacon ; Mario Chacon Rivas ; Cristian Cechinel ; Agustin Francesa Alfaro. Institute of Electrical and Electronics Engineers Inc., 2016.
@inproceedings{107f9971659244dea72e923e553ad145,
title = "Adaptive multilevel clustering model for the prediction of academic risk",
abstract = "The selection of a model for academic risk prediction systems is usually based on the global performance of the model. However, this global performance is not an important factor for the end-user of the system. For the end-user, the performance of the model for his or her specific case is the most important aspect of that model. Given that the model is usually selected at design time, the end-user could end up with a sub-optimal prediction. To solve this problem, this work presents a conceptual framework to build adaptive multilevel clustering models for academic risk prediction. This frameworks allows the system to automatically select between several levels of hierarchical or semi-hierarchical features to create a clustering model to best predict the particular risk of each student. This conceptual framework is validated through its realization into an adaptive model to predict the risk of failing a course during a semester in a Computer Science program. In this study, the adaptive model consistently outperforms the prediction of the best-performing static model.",
author = "Xavier Ochoa",
year = "2016",
month = "11",
day = "21",
doi = "10.1109/LACLO.2016.7751800",
language = "English (US)",
editor = "Chacon, {Isaac Alpizar} and Rivas, {Mario Chacon} and Cristian Cechinel and Alfaro, {Agustin Francesa}",
booktitle = "Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Adaptive multilevel clustering model for the prediction of academic risk

AU - Ochoa, Xavier

PY - 2016/11/21

Y1 - 2016/11/21

N2 - The selection of a model for academic risk prediction systems is usually based on the global performance of the model. However, this global performance is not an important factor for the end-user of the system. For the end-user, the performance of the model for his or her specific case is the most important aspect of that model. Given that the model is usually selected at design time, the end-user could end up with a sub-optimal prediction. To solve this problem, this work presents a conceptual framework to build adaptive multilevel clustering models for academic risk prediction. This frameworks allows the system to automatically select between several levels of hierarchical or semi-hierarchical features to create a clustering model to best predict the particular risk of each student. This conceptual framework is validated through its realization into an adaptive model to predict the risk of failing a course during a semester in a Computer Science program. In this study, the adaptive model consistently outperforms the prediction of the best-performing static model.

AB - The selection of a model for academic risk prediction systems is usually based on the global performance of the model. However, this global performance is not an important factor for the end-user of the system. For the end-user, the performance of the model for his or her specific case is the most important aspect of that model. Given that the model is usually selected at design time, the end-user could end up with a sub-optimal prediction. To solve this problem, this work presents a conceptual framework to build adaptive multilevel clustering models for academic risk prediction. This frameworks allows the system to automatically select between several levels of hierarchical or semi-hierarchical features to create a clustering model to best predict the particular risk of each student. This conceptual framework is validated through its realization into an adaptive model to predict the risk of failing a course during a semester in a Computer Science program. In this study, the adaptive model consistently outperforms the prediction of the best-performing static model.

UR - http://www.scopus.com/inward/record.url?scp=85006856305&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85006856305&partnerID=8YFLogxK

U2 - 10.1109/LACLO.2016.7751800

DO - 10.1109/LACLO.2016.7751800

M3 - Conference contribution

BT - Proceedings - 2016 11th Latin American Conference on Learning Objects and Technology, LACLO 2016

A2 - Chacon, Isaac Alpizar

A2 - Rivas, Mario Chacon

A2 - Cechinel, Cristian

A2 - Alfaro, Agustin Francesa

PB - Institute of Electrical and Electronics Engineers Inc.

ER -