The document presents a novel content sharing scheme called discover-predict-deliver (DPD) for smartphone-based delay-tolerant networks, addressing the inefficiencies of traditional data delivery methods in sporadically connected environments. The approach utilizes mobility learning algorithms and a hidden Markov model to predict future mobility patterns and enhance content discovery and delivery. Evaluation results demonstrate that 87% of content can be successfully shared within 2 hours, with manageable impacts on smartphone CPU and battery life.