SELF-ASSESSMENT ACTIVITIES AS A FACTOR FOR DRIVING THE LEARNING PERFORMANCE: ANALYSIS AND MODELS
SELF-ASSESSMENT ACTIVITIES AS A FACTOR FOR DRIVING THE LEARNING PERFORMANCE: ANALYSIS AND MODELS

Summary/Abstract: Machine learning proposes innovative methods for students' learning analysis and new ways for modeling the learning process and its realization. Learning analytics takes advantages from this fact and processes data according to accepted or emerged algorithms that leads to creation of understandable analytical and predictive models. Learning performance is connected to a set of behavioral activities in educational environment concerning improvement of knowledge and skills. It is a very important criterion for students' progress and for the formation of the final students' outcomes. For achieving better learning performance, the activities should lead to the learning optimization in context of time duration, educational tasks organization, content presentation and management. Activities that support learning are oriented to self-depended and self-regulated learning as well as to socially-oriented and group-driven learning. The aim of the paper is to present an exploration focusing on influence of self-depended activities in the form of self-assessment on learning performance. An experiment is conducted with students who have had possibility to direct and organize their self-assessment activities in learning management system. Self-assessment activities are not graded and they are not included in the formation of the final course mark. The students' behavior is traced during one semester and machine learning algorithms are utilized to analyze the quality and quantity of the taken self-assessment activities. On this base analytical and predictive models regarding learning performance and the achieved academic results are created. The patterns and anomalies are outlined and they are used to point out the directions for learning performance and final outcomes improvement.

• Details
• Contents