Exploring Machine Learning in Higher Education Industry : Student Performance Prediction
Sivula, Ari (2021)
Avaa tiedosto
Lataukset:
2021
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202105189164
https://urn.fi/URN:NBN:fi:amk-202105189164
Tiivistelmä
Machine learning (ML) is utilized constantly in various industries because its possibility to provide innovative solutions to different stakeholders of an organization. ML is utilized in higher education industry to provide insights and supporting activities of an educational institution. The higher education industry organizations have commonly several data sources, which they can adapt in their activities. These systems provide the data, which can be utilized with machine learning algorithms.
The main objective of this study is to explore the ML in higher education industry. Moreover, the objective is to provide an example of ML-based project and its implementation utilizing ML project management approach. The CRISP-DM was selected an approach to implement the development task and answer the research questions. Several supervised and unsupervised ML algorithms were utilized during the research process. The research exists about ML utilization in higher education industries, but each research is conducting a different type of contribution, because of datasets and contexts. This study provides a ML-based solution related to the VIRTA data and systems, which are commonly utilized in a higher education industry organization's in Finland.
The implementation of ML project was a success in overall and the deployment of the models were implemented in this study. The results of this study indicate that CRISP-DM approach can be adapted in higher education industry in several ways and, moreover, machine learning provides value in student performance prediction when appropriate algorithms are developed based on the requirements of an organization and its data. The results of this study can be adapted as well other higher education institutions, which have the VIRTA data. However, more research and data are required to make the student performance prediction more accurate and include more features as well. This additional data could be collected from various systems, for instance, student management, learning management, project management, and reporting systems.
The main objective of this study is to explore the ML in higher education industry. Moreover, the objective is to provide an example of ML-based project and its implementation utilizing ML project management approach. The CRISP-DM was selected an approach to implement the development task and answer the research questions. Several supervised and unsupervised ML algorithms were utilized during the research process. The research exists about ML utilization in higher education industries, but each research is conducting a different type of contribution, because of datasets and contexts. This study provides a ML-based solution related to the VIRTA data and systems, which are commonly utilized in a higher education industry organization's in Finland.
The implementation of ML project was a success in overall and the deployment of the models were implemented in this study. The results of this study indicate that CRISP-DM approach can be adapted in higher education industry in several ways and, moreover, machine learning provides value in student performance prediction when appropriate algorithms are developed based on the requirements of an organization and its data. The results of this study can be adapted as well other higher education institutions, which have the VIRTA data. However, more research and data are required to make the student performance prediction more accurate and include more features as well. This additional data could be collected from various systems, for instance, student management, learning management, project management, and reporting systems.