Education Data Science, new frontier in learning management

The application of machine learning techniques in the field of educational management has marked a pivotal progression in the utilization of data science to redefine and ameliorate educational paradigms. This integration of data science not only facilitates the personalization and inclusivity of learning processes but also ensures that educational practices remain pertinent in a rapidly digitalizing global context. Emphasizing the transformative power of advanced analytics, the study underscores the importance of continuing to harness emerging technologies. This proactive approach to educational innovation not only serves the immediate needs of today's diverse student body but also strategically molds the educational framework for future generations, ensuring it aligns with ongoing technological advancements and societal needs.

At the end of 2023, I presented as the conclusion of my master's degree in digital transformation a work on the analysis and application of machine learning techniques in education management. To this end, I tried to contextualize the changes promoted by recent advances in information technology, both in the view of the role of education and in the issue of "Analytics" allowed by such technologies in educational management. Despite the scarcity of educational theories, we can address two fundamental concepts for the new times: "Intelligent Education" and "Smart Learning". We address advances in the integration of technology and pedagogy, focusing on how these approaches improve the personalization and effectiveness of learning. "Intelligent Education" aims to create a dynamic and inclusive environment that meets the individual needs of students, while "Smart Learning" uses technologies such as artificial intelligence and big data to personalize education.

Main elements of "Intelligent Education":

1. Access to New Technologies: Technology is used to create a dynamic and interactive learning environment.
2. Faculty Preparation: Teachers must know how to use technology effectively to create engaging teaching strategies.
3. Active Learning Environments: Two-way communication between teachers and students is essential for sharing knowledge.
4. Promoting Inclusion and Equity: Adapting accessibility features to fully integrate students with special needs.
5. Self-Learning and Personalization: Learning systems must be able to adjust to the student's individual pace and needs.

Features of "Smart Learning":

1. Use of Advanced Technology: Utilizes AI, big data, and machine learning to shape education based on student needs.
2. Mobile and Ubiquitous Learning: Technology allows learning anywhere and at any time, adapting to the student's context.
3. Context Sensitivity: The ability of systems to learn from the user's context to provide personalized and seamless learning experiences.

Both approaches highlight the importance of an educational design that integrates emerging technologies to provide efficient and adaptive learning, ensuring that all learners have the tools they need to succeed in an increasingly digital and globalized environment.

The fusion between the innovations of "Intelligent Education" and "Smart Learning" with the comprehensive practices of "Educational Data Mining" and "Learning Analytics", which we will detail below, generates an imminent revolution in educational management. By integrating the latest data analytics technologies and machine learning techniques, we can amplify effectiveness and inclusivity in educational processes, while using detailed analytics to tailor educational strategies to individual learners' needs. This synergy between technological advancements and educational data mining enables not only enhanced personalization of teaching, but also offers a perspective of continuous transformation of the educational environment, ensuring that it remains relevant and effective in the digital age. This link between educational technology and data analytics is critical to shaping future generations, equipping them with the tools they need to thrive in an increasingly data-driven world.

Educational Data Mining and Learning Analytics

Educational Data Science (EDS) is an interdisciplinary field that bridges education and data science to analyze and optimize student learning, this science encompasses Educational Data Mining and Learning Analytics . This area has experienced remarkable growth after 2010, attributed to the rise of e-learning and MOOCs between 2008 and 2012. EDS employs techniques such as data analysis, machine learning, and data mining to study student achievement and improve educational quality. Despite its popularity in other sectors, adoption in education has been slower, possibly due to caution about the vast amount of data generated in educational activities. This field shows great potential to improve both teaching effectiveness and student success, taking advantage of the growing volume of available educational data.

Educational data analysis can be applied at various levels, from the classroom to the teaching management level, and can help identify patterns, trends, and opportunities to improve teaching effectiveness. EDS is an area of great potential to improve the quality of education and student success, and its importance has been growing as the amount of education data increases available.

Educational Big Data

Education data encompasses a wide range of information collected from the education system, including data on students, teachers, curricula, and teaching methods, among others. This information is critical to understanding how students learn and retain information, which can significantly improve their readiness for academic and professional success.

Big Data in Education: The expansion of educational data has been notably influenced by the popularity of e-learning and MOOCs, generating a large amount of data that can be analyzed to improve education. The term Big Data refers to extensive data sets that require advanced technologies for processing and analysis, and are characterized by their volume, velocity, and variety.

Big Data Applications

1. Improving Teaching Quality: Through deep data analysis, it is possible to identify trends and patterns that can be used to personalize teaching and improve the educational experience of students.

2. Identification of Needs and Improvements: Data can be used to monitor educational quality, identify risk areas such as dropout, and develop effective education policies.

3. Personalized Support: Testing can detect where students are struggling, allowing for early interventions to offer additional support.

Data Privacy and Security

It is crucial to ensure the protection of student data and establish clear data governance policies to handle information ethically and securely.

The integration of Big Data in the education sector offers numerous opportunities to enrich learning and optimize educational processes, but it also requires careful attention to ethical and privacy aspects.

Educational data is distinctive for involving multiple dimensions of the education system such as student achievement, pedagogical methods, and more. This specificity means that such data is not only varied in its origin and format, but also sensitive, encompassing personal and confidential information that requires careful management.

Classification of Educational Data according to Sampson et al. (2022)

1. Input Data: Related to educational inputs such as student demographics and financial resources.

2. Process Data: Operational information such as the quality of teaching and financial data.

3. Outcome Data: Direct results of educational processes such as dropout rates and performance in exams.

4. Satisfaction Data: Opinions of participants in the educational process, such as students and teachers.

Application and Challenges:

- Education data is a powerful tool for personalizing teaching and improving education policies.

- Given its sensitive nature, it is imperative to implement robust privacy policies to prevent misuse of information.

- The correct interpretation of these data requires a contextualized approach, aware of the educational environment in which they were generated.

Learning Analytics

Learning Analytics (LA) is a field that was formalized in 2011, focused on using data analytics to optimize teaching and learning. It is based on techniques such as machine learning and data mining to analyze student behavior and performance within educational contexts. This area was developed with the idea that, through the effective use of data, it is possible to improve both the educational experience of students and pedagogical methods.

Fundamentals of Learning Analytics:

1. Purpose: To improve decision-making and educational effectiveness by analyzing data on student learning and behavior.

2. Applications: From predictive analytics that forecast future performance to descriptive analytics that provide a retrospective view of student performance. In addition, prescriptive analysis is used to provide concrete recommendations to improve learning.

Differences Between Learning Analytics and Educational Data Mining (EDM)

- EDM: Focuses on the application of data mining techniques to uncover patterns and predict educational outcomes. It is more specific in its approach to data, concentrating on the construction of models that explain or predict student behaviors.

- LA: It has a broader scope that includes not only data mining but also data visualization and statistical analysis to improve real-time educational planning and execution.

The evolution of Learning Analytics has been influenced by the increasing amount of data generated in digital educational environments and the need to personalize education to meet individual student needs.

Reference

McFarland, D. A., Khanna, S., Domingue, B. W., & Pardos, Z. A. (2021). Education data science: Past, present, future. AERA Open, 7. https://doi.org/10.1177/23328584211052055

Sampson, D., Papamitsiou, Z., Ifenthaler, D., Giannakos, M., & Mougiakou, S. (2022). Educational Data Literacy. SPRINGER INTERNATIONAL PU.