Everyone can learn almost about anything online, though learning is not an easy task. A set of modifications happen in the brain when acquiring a new skill, behavior, or knowledge. The attempt to recall information is an integral part of the learning process. The brain reorganizes that information and strengthens that memory like a muscle. The synergy of cognitive psychology and data science have the potential to facilitate learning. We use the Ebbinghaus forgetting curve which is a famous mathematical formula in space repetition models. It hypothesizes that memory retention decays exponentially over time. We explore the Half-Life Regression algorithm - proposed by Settles and Meeder - to predict the recall rates of recent information. It fits the forgetting curve based on real data of students' log traces. Its parameters get estimated and evaluated with gradient descent and regularized squared loss. This algorithm is being used by Duolingo to recommend to learners when and what information to revise to prevent its decay over time.