Procedimiento de agrupación de estudiantes según riesgo de abandono para mejorar la gestión estudiantil en educación superior

Hinojosa, Mauricio; Dérpich, Iván; Alfaro, Miguel; Ruete, David; Caroca, Alejandro; Gatica, Gustavo

doi:10.35699/1983-3652.2022.37275

Cited by 4 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This study is classified as a quantitative research of descriptive type, adapted from other studies [7], widely explained by Sampieri [8], with an extended data mining approach of the Cross-Standard Process for Data Mining in Industry (CRISP-DM) [9] to analyze RCCD and meet the objective of the study. The phases of the process used are described below: For the analysis of information, the Python 3.11.1 programming language, the Apache Spark platform and the PySpark 3.3.2 library were used.…”

Section: Methodsmentioning

confidence: 99%

Analysis of the Public Health Service in Bogotá, Colombia: a Study Based on Customer's Complains and Using Unsupervised Learning Algorithms

Quinchia-Lobo,

Salazar-Gonzalez,

Salas-Álvarez

et al. 2023

Position Papers of the 18th Conference on Computer Science and Intelligence Systems

View full text Add to dashboard Cite

In this study, our aim is to analyze the public health services in the city of Bogota, Colombia. We used unsupervised learning algorithms for clustering requests, complaints, claims, and denunciations issued to Supersalud in 2021. We collected the data from Supersalud's databases. We adopted clustering algorithms such as K-Means, Bisecting K-Means, and Gaussian Mixture, thus, we evaluated the quality of the combination using the silhouette coefficient. The algorithm with the best clustering quality to generate the clusters has been improved. Of the eight clusters, the first two present the highest incidences, with 181 and 249 affiliates affected for every 2,000 in the year 2021. In the first cluster, with 55% support and 100% confidence, a strong association was found between problems related to medical care facilities and restricted access to health services. In addition, in these two clusters RCCD with pathologies such as chronic communicable and non-communicable diseases (respiratory, diabetes, renal, risk factors and cardiovascular) associated with restricted access to health services were found. In conclusion, The unsupervised grouping allowed to analyze the public health services from the perspective of the RCCD, providing valuable information on the experiences of the users and the challenges in the provision of health services in Bogotá, these findings demonstrate the restriction in the access to health services from different perspectives of a deficient state regarding the provision of health services in the city of Bogotá, Colombia.

show abstract

Section: Methodsmentioning

confidence: 99%

Analysis of the Public Health Service in Bogotá, Colombia: a Study Based on Customer's Complains and Using Unsupervised Learning Algorithms

Quinchia-Lobo,

Salazar-Gonzalez,

Salas-Álvarez

et al. 2023

Position Papers of the 18th Conference on Computer Science and Intelligence Systems

View full text Add to dashboard Cite

show abstract

“…compared different computational methods, including machine learning and survival analysis techniques, to predict higher education dropout in various contexts in Latin America Flores et al (2022). analyzed eight predictive models to forecast university dropout in Peru, demonstrating the effectiveness of different analytical approaches Hinojosa et al (2022). identified the profile of students at risk of dropout in Chile to generate student management plans that impact the explanatory variables of dropout Negreiros et al (2021).…”

mentioning

confidence: 99%

Higher Education Dropout: A Scoping Review

Santos,

Pereira,

Pilatti

2024

RGSA

View full text Add to dashboard Cite

Objective: To identify the scope, methodologies, and emerging trends related to dropout in higher education. Method: A scoping review using the PRISMA-ScR methodology, covering articles published between 2019 and 2023 in the databases SciELO, Web of Science, and ScienceDirect. After removing duplicates and applying inclusion and exclusion criteria, 30 articles were selected and critically evaluated to identify patterns and gaps in the literature on higher education dropout. Results and Discussion: The main findings highlight the effectiveness of predictive models and personalized interventions in student retention and the importance of sustainable practices for the longevity of higher education institutions. Numerous factors influence dropout in higher education, and personalized interventions and sustainable practices are essential to increase retention and promote the sustainability of institutions. Research Implications: The research highlights the need for personalized interventions and sustainable practices to reduce dropout, outlining a current framework for researchers in the field and providing elements for managers and policymakers. Originality/Value: The study comprehensively maps the methodologies and emerging trends on higher education dropouts, using a scoping review with methodological rigor and contributing an updated and detailed view of the field.

show abstract

Undergraduate Student Retention Activities: Challenges and Research Agenda

Gonçalves,

Serra,

Storopoli

et al. 2024

Sage Open

View full text Add to dashboard Cite

Our study aims to identify effective activities to enhance student retention and promote student success in higher education institutions (HEIs) by reviewing the factors that affect student retention through a literature review. Our study used a systematic review with the bibliometric method to analyze 125 articles related to the retention of students in higher education from 2014 to 2022. We used PCA and network analysis to identify and describe three research fronts of activities to student retention in higher education, two of which are from the perspective of the higher education institution (HEI) and focus on actions for retention to reduce student dropout and activities developed for student success. The third research front is centered on the student, seeking to understand what can lead to persistence in finishing their course. Our bibliometric study enabled us to identify the works and themes that have influenced current research on retention in higher education. By analyzing the bibliographic coupling component, our study provides insights into the research streams in higher education and highlights areas that require further investigation.

show abstract

Procedimiento de agrupación de estudiantes según riesgo de abandono para mejorar la gestión estudiantil en educación superior

Cited by 4 publications

References 21 publications

Analysis of the Public Health Service in Bogotá, Colombia: a Study Based on Customer's Complains and Using Unsupervised Learning Algorithms

Analysis of the Public Health Service in Bogotá, Colombia: a Study Based on Customer's Complains and Using Unsupervised Learning Algorithms

Higher Education Dropout: A Scoping Review

Undergraduate Student Retention Activities: Challenges and Research Agenda

Contact Info

Product

Resources

About