Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection

"The selection of a suitable master’s program has become an increasingly complex task due to the growing diversity of offerings and the lack of standardized comparable information across institutions. This thesis proposes the development of a recommender system model that utilizes graph databas...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autor principal: Pastores, Claudia
Formato: Tesis de maestría
Lenguaje:Inglés
Publicado: 2025
Materias:
GRAPH DATABASES, RECOMMENDER SYSTEMS, MASTER'S PROGRAM SELECTION, HIGHER EDUCATION, DATA STANDARDIZATON, NEO4J, DECISION SUPPORT
Acceso en línea:https://hdl.handle.net/20.500.14769/5141
Aporte de:
Repositorio Institucional Instituto Tecnológico de Buenos Aires (ITBA) de Instituto Tecnológico de Buenos Aires (ITBA)
id I32-R138-20.500.14769-5141
record_format dspace
spelling I32-R138-20.500.14769-51412026-01-07T14:10:08Z Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection Pastores, Claudia GRAPH DATABASES, RECOMMENDER SYSTEMS, MASTER'S PROGRAM SELECTION, HIGHER EDUCATION, DATA STANDARDIZATON, NEO4J, DECISION SUPPORT "The selection of a suitable master’s program has become an increasingly complex task due to the growing diversity of offerings and the lack of standardized comparable information across institutions. This thesis proposes the development of a recommender system model that utilizes graph database technology to enable a more structured and transparent comparison of postgraduate programs. By representing academic entities (programs, courses and universities) as interconnected nodes, the model supports flexible querying and the visualization of complex academic relationships that are often insufficiently addressed by traditional recommender systems. To facilitate the system’s development, a standardized dataset was compiled by extracting program information from various university websites and official documents using custom web scraping scripts. The implementation of a prototype in Neo4j illustrates the advantages of graph databases in terms of relationship modeling, scalability and adaptability to heterogeneous data structures. The findings contribute to the development of decision-support tools in higher education and offer a foundation for enhancing the comparability and accessibility of academic programs on an international scale." "La selección de un programa de maestría adecuado se ha vuelto una tarea cada vez más compleja debido a la creciente diversidad de opciones y a la falta de información estandarizada y comparable entre instituciones. Esta tesis propone el desarrollo de un modelo de sistema de recomendación que emplea tecnología de bases de datos de grafos con el fin de permitir una comparación más estructurada y transparente de programas de posgrado. Al representar entidades académicas (programas, cursos y universidades) como nodos interconectados, el modelo facilita consultas flexibles y la visualización de relaciones académicas complejas que los sistemas de recomendación tradicionales no logran abordar adecuadamente. Para respaldar el desarrollo del sistema se compiló un conjunto de datos estandarizado mediante la extracción de información desde sitios web y documentos oficiales de diversas universidades utilizando scripts personalizados de web scraping. La implementación de un prototipo en Neo4j demuestra las ventajas de las bases de datos de grafos en cuanto al modelado de relaciones, la escalabilidad y la adaptabilidad a estructuras de datos heterogéneas. Los resultados contribuyen al desarrollo de herramientas de apoyo para la toma de decisiones en la educación superior y sientan las bases para mejorar la comparabilidad y accesibilidad de los programas académicos a nivel internacional." 2025-10-27T13:21:18Z 2025-10-27T13:21:18Z 2025-07-11 Tesis de maestría https://hdl.handle.net/20.500.14769/5141 en application/pdf
institution Instituto Tecnológico de Buenos Aires (ITBA)
institution_str I-32
repository_str R-138
collection Repositorio Institucional Instituto Tecnológico de Buenos Aires (ITBA)
language Inglés
topic GRAPH DATABASES, RECOMMENDER SYSTEMS, MASTER'S PROGRAM SELECTION, HIGHER EDUCATION, DATA STANDARDIZATON, NEO4J, DECISION SUPPORT
spellingShingle GRAPH DATABASES, RECOMMENDER SYSTEMS, MASTER'S PROGRAM SELECTION, HIGHER EDUCATION, DATA STANDARDIZATON, NEO4J, DECISION SUPPORT
Pastores, Claudia
Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
topic_facet GRAPH DATABASES, RECOMMENDER SYSTEMS, MASTER'S PROGRAM SELECTION, HIGHER EDUCATION, DATA STANDARDIZATON, NEO4J, DECISION SUPPORT
description "The selection of a suitable master’s program has become an increasingly complex task due to the growing diversity of offerings and the lack of standardized comparable information across institutions. This thesis proposes the development of a recommender system model that utilizes graph database technology to enable a more structured and transparent comparison of postgraduate programs. By representing academic entities (programs, courses and universities) as interconnected nodes, the model supports flexible querying and the visualization of complex academic relationships that are often insufficiently addressed by traditional recommender systems. To facilitate the system’s development, a standardized dataset was compiled by extracting program information from various university websites and official documents using custom web scraping scripts. The implementation of a prototype in Neo4j illustrates the advantages of graph databases in terms of relationship modeling, scalability and adaptability to heterogeneous data structures. The findings contribute to the development of decision-support tools in higher education and offer a foundation for enhancing the comparability and accessibility of academic programs on an international scale." "La selección de un programa de maestría adecuado se ha vuelto una tarea cada vez más compleja debido a la creciente diversidad de opciones y a la falta de información estandarizada y comparable entre instituciones. Esta tesis propone el desarrollo de un modelo de sistema de recomendación que emplea tecnología de bases de datos de grafos con el fin de permitir una comparación más estructurada y transparente de programas de posgrado. Al representar entidades académicas (programas, cursos y universidades) como nodos interconectados, el modelo facilita consultas flexibles y la visualización de relaciones académicas complejas que los sistemas de recomendación tradicionales no logran abordar adecuadamente. Para respaldar el desarrollo del sistema se compiló un conjunto de datos estandarizado mediante la extracción de información desde sitios web y documentos oficiales de diversas universidades utilizando scripts personalizados de web scraping. La implementación de un prototipo en Neo4j demuestra las ventajas de las bases de datos de grafos en cuanto al modelado de relaciones, la escalabilidad y la adaptabilidad a estructuras de datos heterogéneas. Los resultados contribuyen al desarrollo de herramientas de apoyo para la toma de decisiones en la educación superior y sientan las bases para mejorar la comparabilidad y accesibilidad de los programas académicos a nivel internacional."
format Tesis de maestría
author Pastores, Claudia
author_facet Pastores, Claudia
author_sort Pastores, Claudia
title Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
title_short Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
title_full Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
title_fullStr Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
title_full_unstemmed Design of a Recommender System Model Leveraging Graph Databases for Master's Program Selection
title_sort design of a recommender system model leveraging graph databases for master's program selection
publishDate 2025
url https://hdl.handle.net/20.500.14769/5141
work_keys_str_mv AT pastoresclaudia designofarecommendersystemmodelleveraginggraphdatabasesformastersprogramselection
_version_ 1854267358980341760

Ejemplares similares