Rapid progress in expansion of the internet services have provided an alternative way for learning other than the traditional classroom learning. Due to the availability of multiple learning options, evaluating each option and judging the best use case plays a vital role. One of the most important characteristics that a human brain utilizes during process of learning is cognition that involves attention and retention. Student's attention span and situational interests during learning have always been a subject matter of research. Apart from classroom learning, e-learning (MOOC based learning) is the other most preferred way of learning. Therefore, the objective of this study is to assess attention levels of a learner in MOOC (Massive Open Online Courses) learning environments and compare it with conventional classroom learning using brain signals. The proposed method captures electroencephalogram (EEG) frequency bands of different subjects while going through a short lecture in MOOC/e-learning environment and classroom environment. The captured data points were annotated for attentiveness manually by referring to the subject's feedback and video clips. Machine learning classification model of support vector machines (SVM) was used to classify student's mental state as attentive or nonattentive. Promising results were obtained and experiments revealed that higher attention levels were maintained during MOOC learning environment in comparison to traditional learning approach.
Cross-lingual transfer learning using multilingual models has shown promise for improving performance on natural language processing tasks with limited training data. However, translation can introduce superficial patterns that negatively impact model generalization. This paper evaluates two state-of-the-art multilingual models, Cross-Lingual Model-Robustly Optimized BERT Pretraining Approach (XLM-Roberta) and Multilingual Bi-directional Auto-Regressive Transformer (mBART), on the cross-lingual natural language inference (XNLI) natural language inference task using both original and machine-translated evaluation sets. Our analysis demonstrates that translation can facilitate cross-lingual transfer learning, but maintaining linguistic patterns is critical. The results provide insights into the strengths and limitations of state-of-the-art multilingual natural language processing architectures for cross-lingual understanding.