Integrating Conversational Agents and Learning Analytics in MOOCs (colMOOC)


The main priority tackled in the colMOOC project identifies that there is a need to make MOOCs more collaborative to support teachers and engage and motivate students. Conversational pedagogical agents guide and support student dialogue using natural language both in individual and collaborative settings. Conversational agents have been produced to meet a wide variety of applications and studies exploring the usage of such agents have led to positive results. Integrating this type of conversational agents into MOOCs to trigger peer interaction in discussion groups is expected to considerably increase the engagement and the commitment of online students (and, consequently, reduce MOOCs dropout rate). With the usage of learning analytics more meaningful assessment may be achieved, since educators can evaluate the work’s progress during each step and adapt the following steps accordingly. This also enables learners to reflect on each specific step and ensure they have obtained the necessary knowledge to progress effectively.

Find More about colMOOC

Visit Our Website


The colMOOC project aims to deliver a highly innovative and beyond the current state-of-the-art MOOC model and implementation with the integration of services based on conversational agents and learning analytics. Especially for the CAs, it is emphasized that the latest studies systematically indicate that agent interventions during peer interaction (i.e. online discussions) trigger task relevant cognitive activity that leads to improved learning outcomes at many levels The agent interventions are modeled according to what is known as "Academically Productive Talk" (APT) framework, which essentially refers to modeling the experienced teacher’s “moves” (interventions) during students’ dialogue to make students elaborate in the domain.

The learning experience that this type of agent provides to students is similar to having one more partner in their group trying to respond to this partner’s prompts. The strategic goal of the project is to make a significant contribution towards European universities in their development towards knowledge service providers to the economy and society at large and to target the now transversal competencies of programming, computational and design thinking as well as concrete application of new technologies for teaching.

While the EU universities jointly offer a high level of education, they trail behind in offering novel and transversal competencies as well as in developing joint and novel approaches to MOOC education.

Ultimately, the project aims to provide novel methods for teaching and learning to address the problem of attrition in online learning.


► Aristotle University of Thessaloniki (AUTH), Greece

► Certh, Greece

► Gunet (CU), Greece

► University of Valladolid (UVA), Spain

► Open University of Catalunia (UOC), Spain

► Telefonica Education Digital (TE), Spain

► Saarland University (SU), Germany

► Learnworlds (LW), United Kingdom

► Aarhus University (AU), Denmark

► Enchatted (Associated Partner), Netherlands


Contact Us

Research Areas:
Google Scholar Link:
Personal Webpage:
Conversational Pedagogical Agents Computer Supported Collaborative Learning
+30 2310991938
Kalamaria, Office 22, Aristotle University of Thessaloniki. Thessaloniki, Greece
Google Scholar
Personal Webpage