A key requirement of a Self e-Learning Network is personalisation of learners' access to learning objects by providing results tailored to the individual or group of learners as the response to search queries. Such personalisation requires a User Profile containing information about preferences, aims, and educational history to be stored and used by the system. In this deliverable we review existing learner profile specifications and identify some of their weaknesses. We go on to specify the semantics of the SeLeNe User Profile, together with RDF schemas defining the profile structure. The profile takes some elements from the existing learner profile schemes considered earlier and also adds extra custom elements where existing schemes fail to provide the expressiveness necessary for effective personalisation. We also define structured vocabularies for the specification of learning goals and learning styles within the profile. We discuss how the SeLeNe User Profile could be applied during the search for, and retrieval of, relevant learning objects, and we suggest how a SeLeNe system could present sets of retrieved learning objects as trails --- sequences of learning objects presented in a suggested order of interaction. The proposed trails can be derived from the combination of the user profile with information from the learning object descriptions managed by the SeLeNe system.