In colaps, we focus on studying and modeling user activities and interactions in various learning contexts. In our research, we employ artificial intelligence, machine-learning, and data-mining approaches to model students’ knowledge state and to assess students’ performance. For example, we are interested in exploring the relationship between response times and student performance with the aim to improve the accuracy of predictive student models. Additionally, we are interested in modeling established pedagogies (such as the
Zone of Proximal Development) with the aim to facilitate personalization and adaptation of instruction and feedback. Among others, we explore the use of data to analyze the collaborative construction of artefacts and to model collaborative practice and creative processes.