| With the rapid development of the Internet, wireless devices and mobile computing, the range of information to which a user pays attention is more extensive and the content of information demand and the ways of obtaining information have nothing in common. However, due to the one-size-fits-all interface people have to consume a lot of time and energy to work on the information they need.The challenge, obviously, is not only to make information available to people at any time, at any place, and in any form, but specifically to do the right thing at the right time in the right way.Intelligent user interfaces, because of its adaptivity, context sensitivity and assistance in tasks, improve the efficiency and usability of human-computer interaction, which have become a focus in HCI research areas. The research of the thesis is part of the intelligent user interfaces and its final objective is to make user interface more effective and easier to use and increase user-satisfaction.The adaptive human computer interface proposed by this thesis is a human-machine interface that aims to fit a user or a task by changing its working behaviours. The thesis, a combination of psychology, artificial intelligence and computer technology, studies how the human computer interface dynamically suits a particular user and his or her tasks at hand to meet his or her information requirements perfectly.Aiming at the psychological characteristics of user's information requirements and existing adaptive user interfaces, the thesis has launched a series of research work, and made the following creative achievement.(1) To advance a 4-key-component interaction model based on the psychological characteristics of user's information requirements.This thesis, first of all, analyses the individual psychological characteristics with respect to information requirements on the basis of research in depth of existing interaction models and the cognitive psychology theory. Then the stimulation-reaction scenario of human computer interaction is further discussed. Finally a 4-key-component interaction model is put forward to aim at the construction of user models for adaptive human interface.(2) To propose multi-using patterns oriented user models and novel modeling methods.The web activities in which a user is involved have different property, leading to different using goals or patterns, namely multi-using patterns. To solve them well, the thesis proposes multi-using patterns oriented user models and novel modeling methods. This model, a long-term and dynamic one, can be divided into two layers: one is for group and another for individual. Group models show the users' common property and individual one has not only attributes from group but also private property of his own.The attributes of individual can also be fed back to group as a property.(3) To construct multi-using patterns oriented resource model.To deal with multi-using patterns for a user's information attaining and distribution and heterogeneity of resource on the Internet, this thesis proposes the concept of multi-using patterns oriented resource model and presents the structure and construction of resource model in accordance with the user's background and preference, which is effective to solve the problem of generation of adaptive resources.(4) To build an adaptive mechanism based on the combination of multi-using patterns and MAS technology.Due to the interaction metaphor and features of learning and adaptability, multi-agent technology is embraced to build an adaptive mechanism. On the basis of it, an adaptive human computer interface is designed and implemented. Moreover, the features of interaction and coordination between Agents and how an Agent interacts with his owner are discussed.Finally, the effective work of the prototype has proven that the ideas and methods presented in this paper are correct. With the adaptive human computer interface proposed by this thesis, the user works more efficiently and effectively as well as increases his or her satisf...
|
PDF Full Text Request