This page is out of date! Consult Cloud4all deliverable D204.1 instead.
MyUI (Mainstreaming Accessibility through Synergistic User Modelling and Adaptability) is one of 4 projects building the VUMS (Virtual User Modelling and Simulation Standardisation) cluster. The project aims the mainstreaming of accessibility and personalized ICT-products . Thereby user interfaces will self-adapt to growing individual user models. Implementations for interactive TV devices, digital physiotherapy services and socialisation services will be developed in the projects . User and context information are collected in real-time by an ontology-based context management infrastructure. Empirically based design patterns will be used for the adaption process. The following figure depicts the personalization process aimed in the project:
Uses self-learning and self-adaptive user profiling mechanism to reduce initial effort of interface configuration and profile creation. Most relevant user information will be extracted from the user’s interaction behaviour by analyzing interface events (time-outs, repeated undos, detours). Furthermore, some hardware sensors as eye tracking are also involved in the profiling process. Mechanisms of creating and updating a user profiles are part of the MyUI Context Management Infrastructure . The MyUI adaptation engine rely on empirically based design patterns for specified user and context characteristics. Each pattern is related to specific end user characteristics and context requirements. Accessibility is achieved by combining design patterns that suits for a certain end user and context of use. The MyUI design patterns repository includes six categories of design patterns: Device-specific patterns, Individualization patterns, Interaction patterns, User interface elements, Adaptation rendering patterns and Adaptation dialogue patterns .
Useful information and sources for Cloud4All
- Collecting user and context information in real-time by using an ontology infrastructure .
- Adaptation through design patterns for specified user and context characteristics .
- Self learning user profiling