複合式電子服務之推薦及系統平台之設計

Abstract

企業在網際網路上提供各種電子服務，已是企業電子化之重要趨勢，企業將其服務電子化，轉換成可經由網路取得的商品，以創造新的收入來源與達到更高的效益。由不同的服務提供者所提供的電子服務所組成的複合式電子服務，可以為顧客帶來更大的價值。從流程的觀點來看，複合式電子服務成為電子服務提供者與顧客的一種價值流程，相對的，其所涵蓋的處理流程較複雜，需透過跨組織的電子服務提供者合作完成。許多研究中提出電子服務平台以動態組合與執行複合式電子服務流程。然而，傳統的電子服務平台並沒有支援對複合式電子服務的資料探勘與推薦。 本研究提出一個新的複合式電子服務平台，增加新的功能於傳統的電子服務平台當中。主要包含三項延伸功能：設計電子服務的Metadata、提供表示複合式電子服務的方法、推薦複合式電子服務的流程。本研究使用UML分析語言中的活動圖形表示複合式的電子服務，以ECA規則描述複合式電子服務流程中的控制流。流程中的每個電子服務相對應一個的電子服務選擇規則，包含選擇電子服務提供者的條件於規則中，以XQL描述此規則。並採用UDDI標準，設計適合的屬性以描述電子服務，以作為電子服務之Metadata。在電子服務的Metadata支援下，使得本平台可以提供語義搜尋，並基於電子服務規則，顧客可以動態的選擇電子服務的提供者。最後，本研究將利用關聯規則之資料探勘技術以發掘電子服務發生次數頻繁之條件集合，以及從執行過的複合式電子服務中發掘電子服務之間的順序集合。基於發掘的結果，本系統平台推薦給顧客複合式電子服務流程，以提供顧客在選擇電子服務屬性及複合式電子服務流程之依據。

Providing various e-services on the Internet is becoming an important trend in e-business. Enterprises make e-services available via the Internet to drive new revenue streams or create new efficiencies. Composite e-services, which are composed by various e-services provided by different e-service providers, can be more valuable for customers. From the workflow viewpoint, composite e-services can be viewed as value-added processes for service providers and customers. Composite e-services are complex processes that require cooperation among cross-organizational e-service providers. Several e-service platforms have been proposed to dynamically compose and enact composite e-services. However, conventional e-service platforms do not support the mining and recommendation of composite e-services. This work presents a novel platform capable of supporting e-service metadata, and modeling and recommending composite e-services. Composite e-services are modeled by using the activity diagram, in which ECA (Event/Condition/Action) rules are employed to define the control flow of the composite e-services. Associated with each e-service, an e-service selection rule consisting of predicates expressed in XQL can be defined to select the appropriate e-service provider that satisfies customer needs. Moreover, UDDI is adopted to design appropriate metadata for describing the characteristics of e-services. With the support of e-service metadata, the platform provides semantic search of e-services and dynamic selection of e-service providers based on the defined e-service selection rules. Furthermore, the association rule mining approach is used to discover the frequent predicate sets of e-services and the frequent ordering (flow) sets among the instances of composite e-services. Based on the mining result, the proposed platform provides recommendation of composite e-services to the customer.