Developing an open knowledge discovery support system for a network environment

Abstract

Knowledge Discovery in Databases (KDD) is a highly complex process where a lot of data manipulation tools with different characteristics can, and in fact have to, be used together in an interactive and iterative fashion, to reach the goal of previously unknown, potentially useful information extraction. In this paper we analyze the major sources of complexity in the framework of network organizations, pointing out the necessity to give support to the user in many different ways and at very different levels of granularity, from the use of a single tool, to the management of whole, distributed, KDD projects. Unfortunately, currently available systems lack to support the users in at least some of these features. We then propose a solution based on the Service Oriented Computing paradigm, arguing that the advantages ofthis paradigm, namely openness, modularity, reusability and transparency, as well as ubiquity, can help in the design of an effective support system for Knowledge Discovery in Databases in network environments.

UDDI registry for Knowledge Discovery in Databases services

“…Such proposals are mainly focused on supporting the data mining phase. In order to support users in the design of a whole KDD project, we are working on the development of an open environment based on the service-oriented paradigm, called Knowledge Discovery in Databases Virtual Mart [8] (see Fig. 4).…”

“…These, in turn, are reflected in more intensive information exchange, regarding both final and partial results, as well as data and manipulation tools. In [8,9] we extensively discussed and motivated the use of a Service Oriented Architecture (SOA) in order to build an open, collaborative and distributed environment, where resources can be easily annotated, introduced, accessed, described, composed and activated. There, the general architecture of a platform called Knowledge Discovery in Databases Virtual Mart (KDDVM) is also provided [14].…”

Description of Knowledge Discovery Tools in KDTML

“…Wrapping legacy code is useful to integrate and make interoperable heterogeneous environments. In particular the application we discuss in the following is part of a project for the development of an extensible service oriented KDD support system [3]. To wrap a tool means to build a specific software encapsulating the tool, that can then be viewed and exploited as service.…”

“…In particular, the recent shift towards network organizations, the great variability of tools which can be exploited in a KDD process, and the dynamism of the Data Mining field, where new algorithms and techniques are developed continuously, suggest to view such tools as services to be found and used on the net, in a sort of open-market environment [1], where the user can look for implementations, suggestions, tools evaluation, examples of use etc., and where distributed organizations can share their tools, data and results, thus minimizing re-implementation and enhancing reusability. On the other hand, integration and interoperability issues have to be faced in order to exploit heterogeneous tools developed by different authors, while support has to be given to users for the dynamical discovery of algorithms and data over the Internet [2], [3], [4]. Following the mainstream of the semantic web, the KDD field is thus observing the evolution of XML-based description languages, the definition of domain ontologies [5], [6, chap.…”

“…In particular, in the paper we show the exploitation of the KDTML description for the automatic wrapping of legacy KDD code in a web service environment. This work is part of a more general project for the development of an extensible service oriented Knowledge Discovery support system hereafter called Knowledge Discovery in Databases Virtual Mart (KDDVM) [3].…”