| Summary: | Adaptation and personalisation is aimed at improving the user experience in e-systems. Personalisation was initially applied in the fields of distance learning and web-based educational systems. Adaptation can be also used in e-advertising, to increase customer satisfaction and encourage repeat visits to websites. Several models/frameworks have been designed for adaptation, for instance AHAM, LAOS, AdRosa, and MyAds. Many systems have been developed based on these frameworks. Most previous models/frameworks were primarily designed for personalised educational experience and were aimed at standalone systems, which cannot be (easily) integrated into existing websites in a lightweight manner. In addition, some of them are used in the portal model of advertising, since they match the interests of the publisher and the advertiser. The aim of this work is to overcome the limitations and weaknesses of these models and systems to deliver adaptive advertising. This work firstly attempts to support and facilitate the integration between adaptive systems and business websites. It also introduces a method to control and adapt advertisements located and owned by businesses. This thesis further proposes a generalised model, the Layered Adaptive Advertising Integration (LAAI), as the starting point for the development of an adaptive advertisement system. In a second stage, it presents a study that assesses the effectiveness of a system (AEADS) based on this model, via a trial run of a model prototype with users (both customers and business owners). In a third stage, social networks are used as inputs for the user model of customers, to enhance the efficiency of acquiring user information, as an addition to the user registration process. Furthermore, social interactions, such as the facility to use “like”, are added to the user model, and the delivery process has the ability to apply actions based on this data. Finally, an evaluation of the whole system proposed is conducted, with business owners and Internet users alike.
|
|---|
