Rigorous development of prompting dialogues.

OBJECTIVES: The aim was to support people with cognitive impairment through speech-based dialogues that guide them through everyday tasks such as activities of daily living. The research objectives were to simplify the design of prompting dialogues, to automate the checking of prompting dialogues for syntactic and semantic errors, and to automate the translation of dialogue designs into a form that allows their ready deployment. APPROACH: Prompting dialogues are described using CRESS (Communication Representation Employing Systematic Specification). This is a notation and toolset that allows the flow in a service (such as a dialogue) to be defined in an understandable and graphical way. A dialogue diagram is automatically translated into a formal specification for rigorous verification and validation. Once confidence has been built in the dialogue design, the dialogue diagram is automatically translated into VoiceXML and deployed on a voice platform. RESULTS: All key objectives of the work have been achieved. A variety of significant dialogues have been successfully represented using the CRESS notation. These dialogues have been automatically analysed through formal verification and validation in order to detect anomalies. Finally, the dialogues have been automatically realised on a VoiceXML platform and have been evaluated with volunteer users.

A social science data-fusion tool and the Data Management through e-Social Science (DAMES) infrastructure.

The last two decades have seen substantially increased potential for quantitative social science research. This has been made possible by the significant expansion of publicly available social science datasets, the development of new analytical methodologies, such as microsimulation, and increases in computing power. These rich resources do, however, bring with them substantial challenges associated with organizing and using data. These processes are often referred to as 'data management'. The Data Management through e-Social Science (DAMES) project is working to support activities of data management for social science research. This paper describes the DAMES infrastructure, focusing on the data-fusion process that is central to the project approach. It covers: the background and requirements for provision of resources by DAMES; the use of grid technologies to provide easy-to-use tools and user front-ends for several common social science data-management tasks such as data fusion; the approach taken to solve problems related to data resources and metadata relevant to social science applications; and the implementation of the architecture that has been designed to achieve this infrastructure.

Abstraction and analysis of clinical guidance trees.

OBJECTIVES: The aims of this work were: to define an abstract notation for interactive decision trees; to formally analyse exploration errors in such trees through automated translation to Lotos (language of temporal ordering specification); to generate tree implementations through automated translation for an existing tree viewer, and to demonstrate the approach on healthcare examples created by the CGT (clinical guidance tree) project. APPROACH: An abstract and machine-readable notation was developed for describing clinical guidance trees: Ad/it (abstract decision/interactive trees). A methodology has been designed for creating trees using Ad/it. In particular, tree structure is separated from tree content. Tree structure and flow are designed and evaluated before committing to detailed content of the tree. Software tools have been created to translate Ad/it tree descriptions into Lotos and into CGT Viewer format. These representations support formal analysis and interactive exploration of decision trees. Through automated conversion of existing CGT trees, realistic healthcare applications have been used to validate the approach. RESULTS: All key objectives of the work have been achieved. An abstract notation has been created for decision trees, and is supported by automated translation and analysis. Although healthcare applications have been the main focus to date, the approach is generic and of value in almost any domain where decision trees are useful.