Learning Analytics Using Data-driven Approach
Learning analytics is the measurement, collection, analysis of data from learners and their contexts, which mainly aim to understand and optimize the learning process as well as the learning environments. By leveraging on the collected large-volume data from Smart Learning Partner (SLP) platform at the AICFE, we adopt the data-driven approach to conduct the learning analytics research. It includes building models to predict student’s academic performance, skill and competence on different subjects, and meanwhile analyzing the learning barriers and obstacles on the knowledge maps. Moreover, we also conduct student learning activity recognition, which uses the intelligent sensors and wearable devices to capture and detect different personal activities during the learning process. Multiple machine learning models, such as the artificial neural networks and Bayesian models, are used together with the advanced data analytics techniques. The research results are expected to enable more intelligent online learning systems, especially for our SLP platform, and also bring about influential research impacts.