Educational data mining is widely deployed to extract valuable information and patterns from academic data. This research explores new features that can help predict the future performance of undergraduate students and identify at-risk students early on. It answers some crucial and intuitive questions that are not addressed by previous studies. Most of the existing research is conducted on data from 2-3 years in an absolute grading scheme. We examined the effects of historical academic data of 15 years on predictive modeling. Additionally, we explore the performance of undergraduate students in a relative grading scheme and examine the effects of grades in core courses and initial semesters on future performances. As a pilot study, we analyzed the academic performance of Computer Science university students. Many exciting discoveries were made; the duration and size of the historical data play a significant role in predicting future performance, mainly due to changes in curriculum, faculty, society, and evolving trends. Furthermore, predicting grades in advanced courses based on initial pre-requisite courses is challenging in a relative grading scheme, as students’ performance depends not only on their efforts but also on their peers. In short, educational data mining can come to the rescue by uncovering valuable insights from academic data to predict future performance and identify the critical areas that need significant improvement.

Assistance for doctors in disease detection can be very useful in environments with scarce resources and personnel. Historically, many patients could have been cured with early detection of the disease. The application of deep learning techniques in the fields of medical imaging, on large datasets, has allowed computer algorithms to produce as effective results as medical professionals. To assist doctors, it is essential to have a versatile system that can timely detect multiple diseases in the lungs with high accuracy. Over time, although many classifiers and algorithms have been implemented, however, deep learning models (i.e., CNN, Deep-CNN, and R-CNN) are known to offer better results. After a thorough literature review of the state-of-the-art techniques, this work applies various models such as MobileNet, DenseNet, VGG-16, EfficientNet, Xception, and InceptionV3 to the selected large dataset. The goal is to enhance the accuracy of these algorithms by experimenting with parameter optimizations. We observe that MobileNet produces better results as compared to other models. We implement a deep convolutional GAN to produce synthetic X-ray images containing various pathologies already included in the chosen imbalanced dataset namely NIH Chest X-ray containing 14 classes. The synthetic dataset contains 1193 samples belonging to five classes. We test the suggested model using evaluation measures like recall, precision, and F1-score, along with binary accuracy. The suggested deep learning model produces recall as high as 57%, binary accuracy as 93.4%, F1-Score as 0.553, and AUC as 81. After the inclusion of generated synthetic samples, the value of the F1-score becomes 0.582 resulting in a 5% increase. Though, Generative Adversarial Network (GAN) shows lower performance, however, we encourage further research and experiments to find the versatility of GANs in the field of medical imaging.