Résumé : Trust can be defined as a measure to determine which source of information is reliable and with whom we should share or from whom we should accept information. There are several applications for trust in Online Social Networks (OSNs), including social spammer detection, fake news detection, retweet behaviour detection and recommender systems. Trust prediction is the process of predicting a new trust relation between two users who are not currently connected. In applications of trust, trust relations among users need to be predicted. This process faces many challenges, such as the sparsity of user-specified trust relations, the context-awareness of trust and changes in trust values over time. In this dissertation, we analyse the state-of-the-art in pair-wise trust prediction models in OSNs. We discuss three main challenges in this domain and present novel trust prediction approaches to address them. We first focus on proposing a low-rank representation of users that incorporates users' personality traits as additional information. Then, we propose a set of context-aware trust prediction models. Finally, by considering the time-dependency of trust relations, we propose a dynamic deep trust prediction approach. We design and implement five pair-wise trust prediction approaches and evaluate them with real-world datasets collected from OSNs. The experimental results demonstrate the effectiveness of our approaches compared to other state-of-the-art pair-wise trust prediction models.