Abstract:

Information and Communication Technologies have modified the traditional Higher education institutions and how the teaching-learning process is conducted. Nowadays, most of the universities incorporate different forms of e-learning in their teaching-learning process by using learning management systems, automatic e-assessment tools or more sophisticated learning intelligent systems that benefit from the advances boosted by disciplines as educational data mining and learning analytics.

These systems produce a massive amount of data every day during the learning process of learners. Each learner's interaction within these systems generates a digital trace composed of data, such as navigational data, textual data, the patterns a learner follows when he or she accesses to the available learning resources, etc. Analyzing these rich data sets implies knowing learners better, finding out their profile or even the identification of learning paths to help them to succeed in their learning process. It also makes possible the implementation of intervention measures to reduce dropout, helping to increase retention and success rate on blended and online environments.

The main goal of this paper is to provide an intervention mechanism embedded into a learning intelligent system in order to support learners to succeed in their learning process in online learning environments by using feedback and predictive analytics. This system, named Learning Intelligent System (LIS), is currently under development in the context of a 3-year project at the Universitat Oberta de Catalunya (UOC). In order to fulfill this objective, to begin with, different types of intervention mechanisms will be identified and reviewed, ranging from nudging (or other messaging systems), the use of intelligent conversational agents, to particular human academic support (counseling and mentoring). Following this, an evaluation of these intervention mechanisms will be provided taking into account the requirements and data available at UOC. This evaluation will allow the conceptualization of interventions measures to be implemented in LIS system.

The contribution of this paper is based on the design and application of intervention mechanisms through LIS system which will help teachers, but mainly students to succeed in their online courses. By predicting their success rate using data from the system and their performance level, the system will help and encourage learners to achieve their learning goals, and consequently their courses certification. The findings will contribute, not only improving the teacher-learning process at UOC, but also in many other blended and online environments.

Keywords:

Intervention mechanisms, learning management system, predictive analytics, learning intelligent system, learning analytics, e-assessment, feedback, e-learning.