DMCNet: Diversified model combination network for understanding engagement from video screengrabs

Engagement is an essential indicator of the Quality-of-Learning Experience (QoLE) and plays a major role in developing intelligent educational interfaces. The number people learning through Massively Open Online Courses (MOOCs) other online resources has been increasing rapidly because they provide us with flexibility to learn from anywhere at any time. This provides good experience for students. However, such interface requires ability recognize level engagement students holistic experience. useful both educators alike. understanding challenging task, its subjectivity collect data. In this paper, we propose variety models that have trained on open-source dataset video screengrabs. Our non-deep are based combination popular algorithms as Histogram Oriented Gradient (HOG), Support Vector Machine (SVM), Scale Invariant Feature Transform (SIFT) Speeded Up Robust Features (SURF). deep methods include Densely Connected Convolutional Networks (DenseNet-121), Residual Network (ResNet-18) MobileNetV1. We show performance each using metrics Gini Index, Adjusted F-Measure (AGF), Area Under receiver operating characteristic Curve (AUC). use various dimensionality reduction techniques Principal Component Analysis (PCA) t-Distributed Stochastic Neighbor Embedding (t-SNE) understand distribution data feature sub-space. work will thereby assist obtaining fruitful efficient