Argument selection is considered the essence of the strategy in argumentation-based negotiation. An agent, which is arguing during a negotiation, has t o decide what arguments are the best to persuade th e opponent. In fact, in each negotiation step, the ag ent must select an argument from a set of candidate arguments by applying some selection criterion. For this task, the agent observes some factors of the negotiation context, for instance trust in the oppo nent, expected utility, among others. Usually, argu ment selection mechanisms are defined statically. Howeve r, as the negotiation context varies from a negotiation to another, defining a static selection mechanism it is not useful. For this reason, we pr esent in this paper a novel approach to personalize argum ent selection mechanisms in the context of argumentation-based negotiation. The selection mech anism defines a set of preferences that determine how preferable it is to utter an argument in a give n context. Our approach maintains a hierarchy of preferences in order to learn new preferences and u pdate the existing ones as the agent experience increases. We tested this approach in a simulated multiagent system and obtained promising results.