Nowadays, Small and Medium Enterprises (SMEs) are forced to compete on a global market and to make strategic decisions in short periods of time. In order to allow SMEs access to information technologies which can support their competition on a global scale, public administrations are fostering the setting up of Digital Districts. In this paper, we describe a distributed collaborative data mining platform, named KD-ASP, developed for a Digital District. It is based on the application service provider (ASP) paradigm, which allows SMEs accessing to data mining services over a network and to cut down costs for their acquisition, upgrading and maintenance. KD-ASP allows the users to collaborate on the design of a knowledge discovery process whose execution is then demanded to a workflow engine. Tasks involved in a process are classified as data selection, pre-processing, data transformation, data mining and data visualization, and are made available as Web services.
A KDD Platform based on the Application Service Provider Paradigm
MALERBA, Donato
2008-01-01
Abstract
Nowadays, Small and Medium Enterprises (SMEs) are forced to compete on a global market and to make strategic decisions in short periods of time. In order to allow SMEs access to information technologies which can support their competition on a global scale, public administrations are fostering the setting up of Digital Districts. In this paper, we describe a distributed collaborative data mining platform, named KD-ASP, developed for a Digital District. It is based on the application service provider (ASP) paradigm, which allows SMEs accessing to data mining services over a network and to cut down costs for their acquisition, upgrading and maintenance. KD-ASP allows the users to collaborate on the design of a knowledge discovery process whose execution is then demanded to a workflow engine. Tasks involved in a process are classified as data selection, pre-processing, data transformation, data mining and data visualization, and are made available as Web services.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
