Knowledge Discovery in Services (KDS): Aggregating Software Services to Discover Enterprise Mashups

Abstract

Service mashup is the act of integrating the resulting data of two complementary software services into a common picture. Such an approach is promising with respect to the discovery of new types of knowledge. However, before service mashup routines can be executed, it is necessary to predict which services (of an open repository) are viable candidates. Similar to Knowledge Discovery in Databases (KDD), we introduce the Knowledge Discovery in Services (KDS) process that identifies mashup candidates. In this work, the KDS process is specialized to address a repository of open services that do not contain semantic annotations. In these situations, specialized techniques are required to determine equivalences among open services with reasonable precision. This paper introduces a bottom-up process for KDS that adapts to the environment of services for which it operates. Detailed experiments are discussed that evaluate KDS techniques on an open repository of services from the Internet and on a repository of services created in a controlled environment.

GOAL OF PROJECT

This body of work largely relates to the greater research area of data and information integration. Similar to

the initial notions of Knowledge Discovery in Databases (KDD), our work also considers knowledge discovery but

instead of databases or data mining, we consider new knowledge that can be attained when aggregating complementary

services.

ANALYSIS ON EXISTING SYSTEM

Web 2.0 [24] is a paradigm that overlays the notion of service-oriented computing. In conjunction with the fundamental paradigms, Web 2.0 advocates for the individual user to be the prime stakeholder. Consumer-to-consumer

collaboration technologies and market-oriented environments allow individual users to interact seamlessly. As

such, the individual user may exploit web services in a manner most appropriate to their daily activities. While

business process composition is not a necessary action for the individual end user, the integration of the resulting data

into a common view can be of importance. A service mashup is the simultaneous execution of two or more services to

create an integrated data provision with a more complete description about some object or task. For example, web

services from a web search company such as Google Corporation that provides mapping capability can be

integrated with capabilities from a shipping business such as the United Parcel Service (UPS)

The notion of service mashup currently receives a great deal of attention from academia and industry. Much of the current work involves tools and techniques that instrument the mashup process and subsequently visualize the results

[13], [31]. This body of work largely relates to the greater research area of data and information integration. Similar to

the initial notions of Knowledge Discovery in Databases (KDD), our work also considers knowledge discovery but

instead of databases or data mining, we consider new knowledge that can be attained when aggregating complementary

services

PROBLEM DEFINITION-Disadvantage

There are three major challenges addressed in this work.

1. In environments where web service-based semantic definitions are not available, high-precision syntactical approaches must be in place to infer equivalences among services using direct and indirect information from service specifications (Equivalence Processing).

2. Characteristics that make two of more services capable of integration or mashup (Clustering) must be well understood and adaptable as the nature of service repositories evolve.

3. Of the services that have sufficient equivalence to support integration or mashup, the subset that actually provides value-added information to end users must be identify

IDEA ON PROPOSED SYSTEM

we discuss an approach, which we call Knowledge Discovery in Services (KDS) [3]. KDS is a systematic process for discovering web service candidates for service mashup that may ultimately uncover new knowledge. Within this approach, there is a customized development life cycle that software engineers can use to create new applications based on mashup techniques. Our work also uncovers the aspects of the web service  specifications that are most effective for determining mashup qualification.

The notion of KDS is supported by a second innovation within our work. Interpreting the complementary nature of distributed web services requires the ability to compare and

contrast interface specifications (i.e., input/output messages, operation names, descriptions, service names, etc.). In the broader area of data integration, semantic languages

such as the Resource Definition Framework (RDF) [29] and the Web Ontology Language for Services (OWL-S) [25] have played a significant role. Unfortunately, open services

randomly available over the Internet are, at least currently, not described in terms of semantics. And, even if they use semantics, they do not adhere to a common ontology which

would unify semantics across disparate domains. We introduce enhanced syntactical techniques that subvert these barriers. Although syntactical approaches lack the confidence of semantic approaches, their flexibility are advantageous in open environments. These techniques

are embedded into adaptive software with the capability of analyzing the characteristics of the individual services. The adaptive software attempts to capture human behavior with respect to how software developers name various aspects of the web services that they create. We call this

behavior the developer’s naming tendencies. These tendencies can be codified into rules that inform our adaptive software. Furthermore, in our work, is the inference of

thresholds that govern the sensitivity of our syntactical software. These thresholds are effective in ranking services that are potentially qualified for service mashup and ultimately for KDS

Requirements:

Hardware Requirement:-

                              Hard Disk                               -              20 GB

                              Monitor                                 -              15’ Color with VGI card support

                              RAM                                        -              Minimum 256 MB

                              Processor                              -              Pentium III and Above (or) Equivalent

                              Processor speed                -              Minimum 500 MHz

                                                               

                Software Requirement:-

                              Operating System              -              Windows XP Professional

                              Platform                                -              Visual Studio .Net 2008

                              Database                               -              SQL Server 2005

                              Languages                             -              c#.net