ADELFI: A model for deployment of high performance solutions on Internet/Intranets

• Forschungsstelle: EU FRP
• Projektnummer: 98.0253
• Projekttitel Englisch: ADELFI: A model for deployment of high performance solutions on Internet/Intranets

Schlüsselwörter (Englisch)

Data mining; gene expression; Java; software; graphics; statistics

Alternative Projektnummern (Englisch)

EU project number: 29880

Forschungsprogramme (Englisch)

EU-programme: 4. Frame Research Programme - 1.3 Telematic systems

Kurzbeschreibung (Englisch)

See abstract

Partner und Internationale Organisationen (Englisch)

Imperical College (UK), CHAM (UK), CSM (I), INERIS (F), Logware (D), Wella (D)

Abstract (Englisch)

ADELFI - 'A Model for the Deployment of High Performance Solutions on the Internet and Intranets' aims at developing a client/server software system for data mining.The objective of ADELFI and therein the two major partners, Imperial College (IC) and CHAM of London, to develop generic software tools for the deployment of web-based, computationally intensive (CFD) and data intensive (Data Mining) applications. The general ideas are based on the systems 'Kensington' from IC and 'Phoenics' from CHAM. GeneData's contribution to the project consisted of illustration of applications in gene expression data analysis to the developers, specification of requirements, provision of detailed test cases including sample data, end-user input, test installation and testing.The contract between the project partners ('consortium agreement') finally has been signed in May 2000.After the first year, a datamining prototype of the developed software was available via the internet for first tests on partner side. The first version contained mainly clustering and classification methods with specific visualizers for graphical display of the results. In the respective year this datamining client was subsequently developed under user and partner input. A final version of the software was provided in fall 2000.With beginning of the year 2001 the development of the ADELFI system was almost done and test case evaluations were performed and finished by end-users for data mining applications. According to the EU reviewers these evaluations were carried out quite professionally and reported in the fourth review meeting. In this context they also explicitly expressed their satisfaction with the work done by GeneData. The ADELFI project officially ended end of may 2001, taking the six month extension for the CFD part into account. The general impression of the reviewers from the final review meeting in October 2001 was that all technical objectives have been reached by the consortium members. A solid software development has led to a prototype of an architecture based on a three-tier client/server application model. It is based on the use of the Enterprise Java Beans which is state-of-the-art. Only little additional work should be required to turn the prototype into a commercial product. The prototype has been successfully tested with several applications coming from the scientific and data mining areas. It thus clearly shows its high degree of generality. The consortium has presented a credible exploitation plan and the reviewers were quite impressed by the results already achieved. Concerning the demonstrations, the reviewers were impressed not only due to the functionality that has been demonstrated but also by the adoption of the system by the end-users. The project results are· A middleware software platform based on a three-tier client/server model as an enhancement of Imperial College's existing platform.· Two real applications running on top of the platform: PHOENICS from CHAM and the Imperial College's Data Mining Application.· A service for using these applications on demand validated by real end users in their real working context. In particular CSM will use the service for steel casting, INERIS for atmospheric simulation and simulation of fire in a tunnel, Logware and Wella for analysis of retail business data, and GeneData for comparison of gene or protein expression data.

Datenbankreferenzen (Englisch)

Swiss Database: Euro-DB of the
State Secretariat for Education and Research
Hallwylstrasse 4
CH-3003 Berne, Switzerland
Tel. [phone]
Swiss Project-Number: 98.0253