掌握先機為提升企業競爭力的重要因素之一,在競爭激烈的知識經濟時代,協助企業組織快速又有效地取得產業相關訊息,並提供給組織內各行政部門,作為執行業務的參考依據。資訊檢索(Information Retrieval)或全文檢索(Full-texts Retrieval)為利用關鍵字(Keyword)的方式篩選出所需要的資訊,在資訊過載(Information Overloading)的網際網路時代,不易精確地搜索出符合使用者需要的知識訊息,企業對於知識的擷取、整合、管理、分享與再利用的重要性也相對的提高。本研究開發以語意規則(Semantic Rule)為基礎的新聞分派系統,運用知識本體(Ontology)的明確性表達為職能領域知識(Domain Knowledge)的概念塑模,分析組織概念與職能概念之間的屬性關係,建立職能本體知識庫,再利用語意規則定義出職能與網路新聞之間的關係來解決問題,輔以推論引擎推論出網路新聞所歸屬的組織層級,達成網路新聞分派系統的功能,以協助企業組織內各職能成員能快速的取得相關的網路新聞,提供各職能成員作為執行業務的參考。本研究以A大學的行政組織為例,說明研究問題與系統設計的實作成果,經由實作得知,由知識本體對已知知識的塑模,輔以語意規則的定義,經由推論後可由已知事實的邏輯關係推演出隱含其中的新事實。
Obtaining instant and useful news is one of the important factors to enhance corporate competitiveness. In the era of knowledge-based economy, it is essential for an enterprise to obtain valuable news, and should be classified and delivered to each department efficiently. Information retrieval or full-texts retrieval are web searching engines based on keywords. Due to the rapidly increasing rate of new information and lack of method for comparing and processing, the information overload problems bring the information pollution to the enterprise. How to collect, integrate, identify, classify, deliver and reuse these information are getting important to the enterprise. This paper is developing a semantic rules-based system for news dispatch. The main objective of this research is applying competence ontology to structure domain knowledge, analyzing competence-to-organization attribute relationship, as well as establishing competence ontological knowledge base. Semantic Rules plays a key role to define the relationship between the competence and internet news. Inference engine is a subsidiary implement to infer that which internet news is suitable for which organization unit. The whole system reaches the goal of dispatching news efficiently. This study which takes the executive department of “A University” for example presents the practice results of this semantic rules-based system. This practice demonstrates that new relations can be discovered from the existing information by means of applying knowledge ontology to structure domain knowledge; as well as adding semantic rules to identify information and infer logical relationship.