Business Intelligence is an important aspect of any organized education. With the growing competition and global outlook of the training, metric driven insight has become extremely important for the corporations to analyze, understand and plan their training modules in better ways. Understanding participant’s behavior and their sentiments has been an active area of research over a long time. However, due to practical challenges associated with such adaptations, there hasn’t been significant work towards participant behavior and sentiment tracking in physical classrooms. In this work, we have addressed this significant challenge of the education sector and proposed EMOMETRIC, which is an intelligent system for class that can track participant’s emotion and provide a participant behavioral insight through IOT integrated data intelligence running on Apache Spark Cluster. The proposed system uses model based face and emotion tracking under real use case conditions Results shows that the proposed technique has an overall accuracy of 95% in comparison to the current state of art. The proposed technique also adapts QOS enabled secure MQTT protocol to collect the data by the big data No-SQL storage system. It is also observed that the proposed technique is not only fast and accurate but also illumination and pose invariant. This work can be used as a framework to offer emotion as a service through SAAS platform.
Keywords : Business Intelligence, Participant Behaviour, Emotion, Apache Spark, MQTT Protocol, No-SQL, Big Data.