Various learning pedagogies have been developed and are widely implemented in educational institutions to enhance the learning abilities of students. One popular approach is the Flipped Learning (FL) model, which is adopted by many higher learning institutions worldwide. Flipped learning shifts the traditional classroom model, encouraging students to engage with instructional content outside of class and utilize class time for active learning and discussion. Electroencephalography (EEG) is a non-invasive technique that records brain activity, allowing researchers to study cognitive processes like learning and attention in different aspects which is very effectively measured using Single-channel EEG headsets. Studies have explored various techniques using these EEG signals to enhance the learning ability of students based on the cognitive state of the learner which is highly correlated to the brainwave signal. In this study, the Histogram of Oriented Gradients (HOG) is effectively exploited to extract the key features from brain waves (EEG signal). The standard machine learning classification techniques are exploited to categorize (Weak and Strong) students based on their attention level. This study can help teachers to identify learners who require additional support to improve their learning abilities. This study contributes to the advancement of the flipped classroom model and provides a potential solution for enhancing learning outcomes, especially for underperforming students.