Study On Multi-Domain User Model Oriented Adaptive Web Mapping Service

With the rapid development of digital city in China and the gradual improvement of spatial data infrastructure, spatial data has become increasing diverse and has been updated more frequently than ever.The monotonous, static and isolated spatial information service and traditional data sharing and interoperable pattern will not be in accord with the multiplicate and personalized demand of users. It's urgent to establish the 'fast lane' which can translate spatial data to map. The traditional expert-centric mapping process should be change into user-centric adaptive mapping pattern. By encapsulate the map production process into technical means, the dynamic and personalized map production should be formed.The theory and technology of adaptive web mapping servie are still in the exploring period.As a highly intelligent service, adaptive web mapping service can be realized by the support and coopepersonalization ration among knowledge of multi-fields. Differs from normal individual oriented adaptive cartographic method,this thesis takes multiple domain users as main object of study,focuses on the study of domain user oriented adaptive web mapping service. The achievements are as follows:(1) This thesis has introduced the concept of adaptive theory, summarized the related theories and technology which include user modelling, web mapping geospatial information service and spatial data infrastructure and outlined the key research point of adaptive web mapping service.(2) The thesis has introduced the concept and definition of adaptive map space; proposed the quaternary model of adaptive domain map space based on the classification of group user;presented a multi-dimension data model based on granularity lattice for the demand of the multiple scale and resolution of users;analyzed the different between domain user and individual user presented the characteristic of domain user and the relation of cartographic constraints and user preference;summarized the composition of user knowledge base;discussed the rules,classification,reasoning of knowledge base;proposed multi-domain user modelling method which combines the domain demand and user preference;A web mining method is presented to construct the self-adaptive domain space model and to analyze the user intension behaviors based on the service log documents focusing on the user preferences of spatial scale, feature semantic and feature object. By the analysis of the real service log information verifies the feasibility and effectiveness of this web mining method.(3)The thesis analyzed the architecture of adaptive web mapping service;discussed the design rules of service architecture;proposed the classification of adaptive web mapping service, introduced the fundamental principal of web mapping service;discussed the user oriented multi-agent web mapping service framework in details;A construction method of adaptive processing object is proposed to encapsulate data processing method,enrichment data structure in whole which is the basis of adaptive control due to user model.The lifecycle and communication mechanism of multi-agent is presented in details which can solve the conflict of constraints independently by calculating the synthetic weights of geometry,graphic and semantic. The character of web mapping service chain is introduced and well defined.The service composition method based on multi-agent is processing to control the map generalization by constraints. Taking user preference into consideration, a optimization method of service composition is processing and discussed in details.(4) A prototype of adaptive web mapping system is designed and implemented based on the research results of this thesis.The main component include user information gathering,user modelling,serving management and online web mapping is introduced.Some strategies and instances is presented to show the process of adaptive web mapping.Discussions mainly focused on the adaptive strategies in user preference,feature relational preference,domain demand,color symbols,thematic service and map style,also the relational examples have been given.