It has emerged from the previous chapters that MAIS was conceived to fulfil user needs through adaptation to the context and personalization. Services that adapt and permit personalization can be conceived such that they take into account different levels of user needs and preferences, such as those relating to devices, quality of service, and visualization. Services can also take into account the context of use and the specific properties of business, in order to enable adoption in service-oriented business environments. Service-oriented architectures are, at the moment, the most promising paradigm for business middleware. This trend is confirmed by the interest of many organizations and standardization bodies involved in the promotion of diffusion such as OASIS, UN/CEFACT and the Value Chain Group. In order to create seamless and fluid business environments, in which people can conduct business as they normally do in the business context, it is necessary to develop systems that enable strong business personalization, capable of delivering business services to users in accordance with their user profile. In this chapter, we shall define a software component, called the recommendation environment, that extends service personalization by enabling a matchmaking process on nonfunctional, semantically rich user and service descriptions, and describe the modeling and design of it. The recommendation environment adds value to the MAIS platform, adding a business dimension with which it is possible to describe e-services, and that at the same time enables reasoning, knowledge extraction, and management. The recommendation environment is placed in the back-end architecture of the MAIS platform (see Fig. 2.3); its role is to recommend, once the functional selection has been carried out, the most suitable concrete e-service with respect to a user profile. Starting from a set of functionally equivalent concrete e-services, the recommendation environment will state which is the service closest to the behavioral description of a user profile. The user profile contains properties that extend the technological description of an e-service by defining, for example, the characteristics of the real world products and services delivered by the company that manages this specific e-service. In order for the recommendation environment to perform its tasks, it is necessary to provide the environment with a back-office tool for data mining capable of analyzing and extracting knowledge from business events generated by the MAIS platform. Starting from basic definitions, in the rest of this section we introduce the concept of a recommender system and describe how recommender systems can support service-oriented architectures in emerging e-business models. In Section 10.2, we describe the architecture of the recommendation environment, specifying its role in the MAIS architecture and its interaction with other components. In Section 10.3, we show how a recommendation is performed, and describe the approach followed for description of users and services, and the algorithm used to evaluate the degree of affinity between a user profile and a concrete e-service. Finally, in Section 10.4, we provide some details about data mining and behavioral profiling, which supports the creation and management of profiles.

Knowledge-Based Tools for E-Service Profiling and Mining

CORALLO, Angelo;LORENZO, Gianluca;SOLAZZO, GIANLUCA;
2006-01-01

Abstract

It has emerged from the previous chapters that MAIS was conceived to fulfil user needs through adaptation to the context and personalization. Services that adapt and permit personalization can be conceived such that they take into account different levels of user needs and preferences, such as those relating to devices, quality of service, and visualization. Services can also take into account the context of use and the specific properties of business, in order to enable adoption in service-oriented business environments. Service-oriented architectures are, at the moment, the most promising paradigm for business middleware. This trend is confirmed by the interest of many organizations and standardization bodies involved in the promotion of diffusion such as OASIS, UN/CEFACT and the Value Chain Group. In order to create seamless and fluid business environments, in which people can conduct business as they normally do in the business context, it is necessary to develop systems that enable strong business personalization, capable of delivering business services to users in accordance with their user profile. In this chapter, we shall define a software component, called the recommendation environment, that extends service personalization by enabling a matchmaking process on nonfunctional, semantically rich user and service descriptions, and describe the modeling and design of it. The recommendation environment adds value to the MAIS platform, adding a business dimension with which it is possible to describe e-services, and that at the same time enables reasoning, knowledge extraction, and management. The recommendation environment is placed in the back-end architecture of the MAIS platform (see Fig. 2.3); its role is to recommend, once the functional selection has been carried out, the most suitable concrete e-service with respect to a user profile. Starting from a set of functionally equivalent concrete e-services, the recommendation environment will state which is the service closest to the behavioral description of a user profile. The user profile contains properties that extend the technological description of an e-service by defining, for example, the characteristics of the real world products and services delivered by the company that manages this specific e-service. In order for the recommendation environment to perform its tasks, it is necessary to provide the environment with a back-office tool for data mining capable of analyzing and extracting knowledge from business events generated by the MAIS platform. Starting from basic definitions, in the rest of this section we introduce the concept of a recommender system and describe how recommender systems can support service-oriented architectures in emerging e-business models. In Section 10.2, we describe the architecture of the recommendation environment, specifying its role in the MAIS architecture and its interaction with other components. In Section 10.3, we show how a recommendation is performed, and describe the approach followed for description of users and services, and the algorithm used to evaluate the degree of affinity between a user profile and a concrete e-service. Finally, in Section 10.4, we provide some details about data mining and behavioral profiling, which supports the creation and management of profiles.
3540310061
978-354031006-8
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11587/113295
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 5
  • ???jsp.display-item.citation.isi??? ND
social impact