Experimentation of proactive computing in context aware systems: Case study of human-computer interactions in e-learning environment

In this paper we show the implementation of the concept of Proactivity applied as the core mechanism in our Proactive Context Aware System (PCAS), which is capable to detect and extract the related events of interest from the user's contextual situation and to provide the appropriate goal-oriented actions to this event with the objective to help or assist the user or the group of users. We have chosen the academic environment as the ongoing contextual setting for our system. From this perspective, we designed Proactive Scenarios for an automatic and enhanced management of the online learning and teaching activities on Moodle™ for both student and teacher users. Due to the diversity of the potential contexts and situations arising from the user's activity, we developed two kinds of Proactive Scenarios. The first type or Meta Scenarios are responsible for capturing the changes in the outward context. The second type or Target Scenarios are triggered off by Meta Scenarios and aim to undertake the appropriate actions in response to the conditions of the user's contextual situation. In order to test and validate the capability of our software as well as to analyse its context related outcomes we have performed empirical studies. The experiments consisted in creating two groups of students, on the one hand the study group, which used the Moodle™ platform enhanced by PCAS, and on the other hand the control group, which used the standard version of Moodle™. The subsequent data analysis showed significant differences in specific related results such as notable advantage of the study group outcomes in the category of passing the final exam, where the study group has performed by 11 percentage points better than the control group.