The proposed project aims to develop an extended statistical information system for commercial statistical databases, capable of guiding the user in the selection of the most appropriate data analysis method while proceed with some degree of interpretation of the data analysis results. It will also help users describe the attributes that characterise the variables of the data set through a wizard in order to allow the system to suggest the appropriate data analysis method and will allow classification of output parameters in order to be used as input to other methods. Further it will work, as regular statistical software with any data set not confined in specific databases. The main advantage of the system is that the user does not need to be an expert i statistics since data analysis and interpretation of the results would be performed automatically where possible.
The proposed project aims at the development of a user-friendly prototype, which will help the users to perform analysis of statistical data. This data analysis tool will also assist the users to derive useful conclusions from the statistical data processed.
Under this project's framework an extended statistical information system will be developed which will provide the:
- Ability to use data analysis methodologies followed by knowledge of experts in the statistical field, by non-experts.
- Ability to proceed with interpretation of the generated results.
- Prototype system able to work as a valuable tool to assist data analysis of commercial statistical databases.
- Ability of the system to work as regular statistical software with any data set not confined in specific databases.
- Creation of a standard interface (API) for the statistical methods that will allow classification of output parameters in order for the output to be used as input to other methods and expandability by third party modules.
The proposed system is targeted at covering the needs both of data providers selling data and of the end-users dealing with data analysis.
The main concept behind X-STATIS is that the system will be used by novice data analysts, therefore allowing guidance in an intelligent way, portraying at any time the overall structure of a data analysis session, and interpreting the data analysis results. It is essential for the proposed system to maximise the data analyst's productivity and satisfaction by providing the analyst not only with the correct guidance but also with tools to expand the system in a suitable data analysis environment capable of conforming to continuously changed user needs. Integrated with a statistical database, it will provide an environment for data analysis regardless of the user sophistication without sacrificing the strengths of standard statistical system features that have proved useful over the years. X-STATIS is thought to accommodate for system intelligence, coupled with powerful data analysis methods and tools to provide extensibility and adaptation.
A step by step approach for the realisation of the project objective will include:
- Research and study of the statistical software available today in the European market and capturing user requirements.
- Study how methodologies used by experts can be adapted and transferred by the system to non-experts.
- Use results obtained from previous or under progress projects.
- Develop and implement the proposed system.
- Verify the system at selected test bed sites, collect feedback, and perform improvements.
- Disseminate project concepts.
For the successful implementation of the aforementioned steps, a project work-plan has been designed. The work to be undertaken has been broken down into the following seven work-packages and the corresponding tasks:
- WP1. Project Management
- WP2. User Requirements
- WP3. Functional Specificaions
- WP4. System Development
- WP5. Validation
- WP6. Exploitation & Dissimination
- WP7: Assesment & Evaluation
Throughout the implementation of the project several results are expected to be achieved. The successful implementation of the project will be measured by the achievement of some predefined milestones.
The most important ones are listed next:
- Accurate capture of the user requirements;
- Development of a user-friendly statistical environment, which will fulfil and satisfy all the predefined user requirements;
- Consolidation of the achieved results and gain of the acceptance of the users; Enlargement of the potential users' basis; Assuring of the commercial viability of the system.
Funding SchemeCSC - Cost-sharing contracts
SW1V 2QQ London