Brief Review of Educational Applications Using Data Mining and Machine Learning

Abstract

The large amounts of data used nowadays have motivated research and development in different disciplines in order to extract useful information with a view to analyzing it to solve difficult problems. Data mining and machine learning are two computing disciplines that enable analysis of huge data sets in an automated manner. In this paper, we give an overview of several applications using these disciplines in education, particularly those that use some of the most successful methods in the machine learning community, such as artificial neural networks, decision trees, Bayesian learning and instance-based methods. Although these two areas of artificial intelligence have been applied in many real-world problems in different fields, such as astronomy, medicine, and robotics, their application in education is relatively new. The search was performed mainly on databases such as EBSCO, Elsevier, Google Scholar, IEEEXplore and ACM. We hope to provide a useful resource for the education community by presenting this review of approaches.Keywords: Education, Data mining, Machine learning. ResumenLa gran cantidad de datos utilizados en la actualidad han motivado la investigación y el desarrollo en diferentes disciplinas buscando extraer información útil con el fin de analizarla para resolver problemas difíciles. La Minería de datos y el Aprendizaje automático son dos disciplinas informáticas que permiten analizar enormes conjuntos de datos de forma automática. En este documento proporcionamos un panorama de varias aplicaciones que utilizan estas disciplinas en la Educación, particularmente aquellas que utilizan algunos de los métodos más exitosos en la comunidad de aprendizaje automático, como redes neuronales artificiales, árboles de decisión, aprendizaje bayesiano y métodos basados en instancias. Aunque estas dos áreas de la inteligencia artificial se han aplicado en muchos problemas del mundo real en diferentes campos, como la Astronomía, la Medicina y la Robótica, su aplicación en la Educación es relativamente nueva. La búsqueda se realizó principalmente en bases de datos como EBSCO, Elsevier, Brief review of educational applications using data mining and machine learning Google Scholar, IEEEXplore y ACM. Esperamos proporcionar un recurso útil para la comunidad educativa con esta revisión de enfoques.Palabras clave: Educación, Minería de datos, Aprendizaje automático.

Machine Learning in Educational Sciences

“…ML-based applications for supporting individual learning requirements or grading students' performance have already entered the classrooms (see, Nájera & de la Calleja Mora, 2017). The application of individualized MLsupported feedback and the inclusion of new learning environments (see, Ciolacu et al, 2018) are a reality that must be considered in the scientific view of a contemporary classroom situation.…”

“…Also, in the present review we extend beyond the topic of mere data mining in education, which has triggered several interesting reviews already (e.g., Ali, 2013;Guleria & Sood, 2014;Romero & Ventura, 2020). Similarly, we limit the discussion of ML-based tools for the improvement of education to particular aspects, relevant to the central message of this review, as broad reviews with a focus on this singular topic have been published (e.g., Kučak et al, 2018;Nájera & de la Calleja Mora, 2017).…”

Using Machine Learning Techniques to Earlier Predict Student's Performance

“…It happen because of the increase of educational resources and data that can be explored to learn how a student learned [17]. Researchers also investigated the factors that influence learning outcomes [18,19,20].…”

Grouping Medical Record Data By Type Diseases With K-Means Algorithm

“…With the data mining technique, data that initially seems unimportant can be used to find useful information from a very large data set with the aim of getting a decision that is very easy to implement and get good and accurate results [4], [5]. Data mining is a process that uses statistical, mathematical, artificial intelligence, and machine learning techniques that extract and identify useful information and knowledge that is assembled from various large databases [6], [7] [8].…”