An intelligent form system.

This thesis presents an investigation of developing a user-centred formbase system. It is based on the previous developments in Office Information Systems. It is technically related to AI Planning Systems and Database Systems. An Office System is an open system inside which the data, as well as the operations upon the data can not be pre-defined exactly. In order to set up a stable and flexible information system in such environments, the task representation, activity representation and data representation must be dynamically related to each other. This research concerns how to use AI planning system concepts to develop a formbase system. There are three crucial aspects: (a) how to represent an activity of information processing, (b) how to represent and refer to the data in the forms, and (c) how to construct the problem solving process for the task of information processing. For reasons of flexibility and stability in the open environment, it is important that a proper link between data representation and activity representation is achieved. This research has generated an Intelligent Form System. The contributions are: (a) the development of a form pattern language and the formbase which can represent and refer to the forms, (b) the identification of the formbase activity schema which can represent the activity upon the forms, and (c) the development of a problem solving process for the information processing tasks of the forms. The research has also recognized that the information processing activities upon forms are very different from the activities which are automatically performed by the Humans. Key Words: Office Information Systems; AI Planning Systems; Activity Representation; Action Reasoning; Knowledge Representation; Office Form Systems.