Digital learning spaces are increasingly generating various forms of data traces (structured, semi-structured, and unstructured). The application of artificial intelligence techniques (AI), such as machine learning, to the analysis of these new forms of data, can offer researchers opportunities for understanding complex human learning. Moreover, the application of statistical analyses together with machine learning and other AI techniques to student data can help teachers gain useful insight into the relationships between student behaviours and outcomes. These techniques can also be used to support assessment. The provided insight or assessment results can then be used to inform the design and support of teaching and learning experiences.
Outside of a specific type of adaptive learning environment called an intelligent tutoring system, the application of machine learning and learning analytics to student data has not progressed to offer students individual, self-directed, and adaptive learning through other platforms. When these methods have been used, they have generally supported the adaptation of learning experiences or the adaptation of exercises that are used to jointly support learning and assessment. This adaptation is usually performed by technologies in domains where the correctness of answers can be determined (e.g., mathematics, physics, verb conjugations). The increasingly large educational data sets that are now being collected enable us to take an individualized approach in a broader set of technologies and domains, but only if we can appropriately automate the analysis of this data and use it to support machine or human decision-making (an adaptation of learning).
This Research Topic aims to bring together research work from various disciplines on the use of analytic techniques to provide individualized learning experiences in formal or informal learning settings.
Topics may include, but are not limited to:
• The use of Data Science approaches in education (e.g., machine learning) or the use of AI, analytics, and data mining to inform adaptation (by a human or machine) in any learning context. Potential examples of the methods which might be used include:
o Social network Analysis
o Sentiment Analysis
o Clustering
o Association rule mining
o Classification (e.g., random forest, decision trees, Naive Bayes, SVM)
o Regression
o Neural Networks
o Coh-Metrix
o Word embeddings (e.g., BERT, Word2Vec)
• The use of data science approaches to perform, improve, or individualize assessment. Potential application areas include:
o Automated essay scoring
o Oral fluency assessment
o Determining communicative competence
• The expansion of analytics designed for adapting learning experiences to a single student so that those analytics can now be used to support a group of students
• The communication of analysis results to teachers and how they use the analyses to inform the teaching and learning environment
Deadlines:
Abstract submission: no earlier than December 15th 2019
Scoping decisions: January 30th 2020
Paper Submission: June 15th 2020
Reviews out: September 30th 2020
Revised submissions: December 21st 2020
Decisions: March 15th 2021
Digital learning spaces are increasingly generating various forms of data traces (structured, semi-structured, and unstructured). The application of artificial intelligence techniques (AI), such as machine learning, to the analysis of these new forms of data, can offer researchers opportunities for understanding complex human learning. Moreover, the application of statistical analyses together with machine learning and other AI techniques to student data can help teachers gain useful insight into the relationships between student behaviours and outcomes. These techniques can also be used to support assessment. The provided insight or assessment results can then be used to inform the design and support of teaching and learning experiences.
Outside of a specific type of adaptive learning environment called an intelligent tutoring system, the application of machine learning and learning analytics to student data has not progressed to offer students individual, self-directed, and adaptive learning through other platforms. When these methods have been used, they have generally supported the adaptation of learning experiences or the adaptation of exercises that are used to jointly support learning and assessment. This adaptation is usually performed by technologies in domains where the correctness of answers can be determined (e.g., mathematics, physics, verb conjugations). The increasingly large educational data sets that are now being collected enable us to take an individualized approach in a broader set of technologies and domains, but only if we can appropriately automate the analysis of this data and use it to support machine or human decision-making (an adaptation of learning).
This Research Topic aims to bring together research work from various disciplines on the use of analytic techniques to provide individualized learning experiences in formal or informal learning settings.
Topics may include, but are not limited to:
• The use of Data Science approaches in education (e.g., machine learning) or the use of AI, analytics, and data mining to inform adaptation (by a human or machine) in any learning context. Potential examples of the methods which might be used include:
o Social network Analysis
o Sentiment Analysis
o Clustering
o Association rule mining
o Classification (e.g., random forest, decision trees, Naive Bayes, SVM)
o Regression
o Neural Networks
o Coh-Metrix
o Word embeddings (e.g., BERT, Word2Vec)
• The use of data science approaches to perform, improve, or individualize assessment. Potential application areas include:
o Automated essay scoring
o Oral fluency assessment
o Determining communicative competence
• The expansion of analytics designed for adapting learning experiences to a single student so that those analytics can now be used to support a group of students
• The communication of analysis results to teachers and how they use the analyses to inform the teaching and learning environment
Deadlines:
Abstract submission: no earlier than December 15th 2019
Scoping decisions: January 30th 2020
Paper Submission: June 15th 2020
Reviews out: September 30th 2020
Revised submissions: December 21st 2020
Decisions: March 15th 2021