The research reported in this thesis proposes, designs and tests a novel Quality of Experience Layer (QoE-layer) for the classic Adaptive Hypermedia Systems (AHS) architecture. Its goal is to improve the end-user perceived Quality of Service in different operational environments suitable for residential users. While the AHS’ main role of delivering personalised content is not altered, its functionality and performance is improved and thus the user satisfaction with the service provided. The QoE Layer takes into account multiple factors that affect Quality of Experience (QoE), such as Web components and network connection. It uses a novel Perceived Performance Model that takes into consideration a variety of performance metrics, in order to learn about the Web user operational environment characteristics, about changes in network connection and the consequences of these changes on the user’s quality of experience. This model also considers the user’s subjective opinion about his/her QoE, increasing its effectiveness and suggests strategies for tailoring Web content in order to improve QoE. The user related information is modelled using a stereotype-based technique that makes use of probability and distribution theory. The QoE-Layer has been assessed through both simulations and qualitative evaluation in the educational area (mainly distance learning), when users interact with the system in a low bit rate operational environment. The simulations have assessed “learning” and “adaptability” behaviour of the proposed layer in different and variable home connections when a learning task is performed. The correctness of Perceived Performance Model (PPM) suggestions, access time of the learning process and quantity of transmitted data were analysed. The results show that the QoE layer significantly improves the performance in terms of the access time of the learning process with a reduction in the quantity of data sent by using image compression and/or elimination. A visual quality assessment confirmed that this image quality reduction does not significantly affect the viewers’ perceived quality that was close to “good” perceptual level. For qualitative evaluation the QoE layer has been deployed on the open-source AHA! system. The goal of this evaluation was to compare the learning outcome, system usability and user satisfaction when AHA! and QoE-ware AHA systems were used. The assessment was performed in terms of learner achievement, learning performance and usability assessment. The results indicate that QoE-aware AHA system did not affect the learning outcome (the students have similar-learning achievements) but the learning performance was improved in terms of study time. Most significantly, QoE-aware AHA provides an important improvement in system usability as indicated by users’ opinion about their satisfaction related to QoE.