Gismate

by O'Brien, James Anthony.

iii Broadly speaking, the aims of this dissertation are to examine the potential of new computational representation techniques and reasoning systems to (i) track various aspects of semantics as information products are created, modified and applied and (ii) utilize stored semantic knowledge to generate the workflows or query execution plans that will produce the required information products in response to specific user-requests. In order to solve the problems outlined above, the feasibility of a framework capable of incorporating semantic and syntactic descriptions of resources and of searching for knowledge that can help solve a specific posed problem is investigated. Thus a hypothesis is presented that it is possible to identify automatically potentially useful resources and to employ these resources effectively to solve a user’s problem. The framework investigated in this dissertation (referred to as GISmate) focuses on each of the resources involved (data, methods, and human experts) and the roles they each play in the evolution of a new information product. GISmate uses a reasoning engine to construct and select the range of possible solutions using a variety of resources in different combinations. A single graph representing the optimal (least cost) solution is extracted from these possible solutions, and displayed in real-time. The description produced represents the formation history of each new information product (such as a map or overlay) and can be stored, indexed and searched as required. Proof of concept examples are presented to show (1) the construction and visualization of information products, (2) the reasoning capabilities of the system to find alternative ways to produce information products from a set of data, methods and expertise, given certain constraints, (3) the construction of a query path to produce a given result and (4) the representation of the ensuing semantic changes by which an information product is synthesized.